Deep Models

Models

class cca_zoo.deepmodels.DCCA(latent_dims: int, objective=<class 'cca_zoo.deepmodels.objectives.MCCA'>, encoders=None, r: float = 0, eps: float = 1e-05, **kwargs)

Bases: _BaseDeep, _BaseCCA

A class used to fit a DCCA model.

References

Andrew, Galen, et al. “Deep canonical correlation analysis.” International conference on machine learning. PMLR, 2013.

forward(views, **kwargs)

We use the forward model to define the transformation of views to the latent space :param views: batches for each view separated by commas

loss(views, **kwargs)

Loss for model :param views: :param args: :param kwargs: :return:

pairwise_correlations(loader: DataLoader, train=False)

Calculates correlation for entire batch from dataloader

Parameters

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

Returns

by default returns the average pairwise correlation in each dimension (for 2 views just the correlation)

score(loader: DataLoader, **kwargs)

Returns average correlation in each dimension (averages over all pairs for multiview)

Parameters

• **kwargs –

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

configure_callbacks()

Configure model-specific callbacks. When the model gets attached, e.g., when .fit() or .test() gets called, the list or a callback returned here will be merged with the list of callbacks passed to the Trainer’s callbacks argument. If a callback returned here has the same type as one or several callbacks already present in the Trainer’s callbacks list, it will take priority and replace them. In addition, Lightning will make sure ModelCheckpoint callbacks run last.

Returns

A callback or a list of callbacks which will extend the list of callbacks in the Trainer.

Example:

def configure_callbacks(self):
    early_stop = EarlyStopping(monitor="val_acc", mode="max")
    checkpoint = ModelCheckpoint(monitor="val_loss")
    return [early_stop, checkpoint]

add_module(name: str, module: Optional[Module]) → None

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Parameters

• name (str) – name of the child module. The child module can be accessed from this module using the given name

• module (Module) – child module to be added to the module.

all_gather(data: Union[Tensor, Dict, List, Tuple], group: Optional[Any] = None, sync_grads: bool = False) → Union[Tensor, Dict, List, Tuple]

Allows users to call self.all_gather() from the LightningModule, thus making the all_gather operation accelerator agnostic. all_gather is a function provided by accelerators to gather a tensor from several distributed processes.

Parameters

• data – int, float, tensor of shape (batch, …), or a (possibly nested) collection thereof.

• group – the process group to gather results from. Defaults to all processes (world)

• sync_grads – flag that allows users to synchronize gradients for the all_gather operation

Returns

A tensor of shape (world_size, batch, …), or if the input was a collection the output will also be a collection with tensors of this shape.

apply(fn: Callable[[Module], None]) → T

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also torch.nn.init).

Parameters

fn (Module -> None) – function to be applied to each submodule

Returns

self

Return type

Module

Example:

>>> @torch.no_grad()
>>> def init_weights(m):
>>>     print(m)
>>>     if type(m) == nn.Linear:
>>>         m.weight.fill_(1.0)
>>>         print(m.weight)
>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

property automatic_optimization: bool

If set to False you are responsible for calling .backward(), .step(), .zero_grad().

backward(loss: Tensor, optimizer: Optional[Steppable], optimizer_idx: Optional[int], *args: Any, **kwargs: Any) → None

Called to perform backward on the loss returned in training_step(). Override this hook with your own implementation if you need to.

Parameters

• loss – The loss tensor returned by training_step(). If gradient accumulation is used, the loss here holds the normalized value (scaled by 1 / accumulation steps).

• optimizer – Current optimizer being used. None if using manual optimization.

• optimizer_idx – Index of the current optimizer being used. None if using manual optimization.

Example:

def backward(self, loss, optimizer, optimizer_idx):
    loss.backward()

bfloat16() → T

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns

self

Return type

Module

buffers(recurse: bool = True) → Iterator[Tensor]

Returns an iterator over module buffers.

Parameters

recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

torch.Tensor – module buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for buf in model.buffers():
>>>     print(type(buf), buf.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

children() → Iterator[Module]

Returns an iterator over immediate children modules.

Yields

Module – a child module

clip_gradients(optimizer: Optimizer, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Handles gradient clipping internally.

Note

Do not override this method. If you want to customize gradient clipping, consider using configure_gradient_clipping() method.

Parameters

• optimizer – Current optimizer being used.

• gradient_clip_val – The value at which to clip gradients.

• gradient_clip_algorithm – The gradient clipping algorithm to use. Pass gradient_clip_algorithm="value" to clip by value, and gradient_clip_algorithm="norm" to clip by norm.

configure_gradient_clipping(optimizer: Optimizer, optimizer_idx: int, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Perform gradient clipping for the optimizer parameters. Called before optimizer_step().

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

• gradient_clip_val – The value at which to clip gradients. By default value passed in Trainer will be available here.

• gradient_clip_algorithm – The gradient clipping algorithm to use. By default value passed in Trainer will be available here.

Example:

# Perform gradient clipping on gradients associated with discriminator (optimizer_idx=1) in GAN
def configure_gradient_clipping(self, optimizer, optimizer_idx, gradient_clip_val, gradient_clip_algorithm):
    if optimizer_idx == 1:
        # Lightning will handle the gradient clipping
        self.clip_gradients(
            optimizer,
            gradient_clip_val=gradient_clip_val,
            gradient_clip_algorithm=gradient_clip_algorithm
        )
    else:
        # implement your own custom logic to clip gradients for generator (optimizer_idx=0)

configure_optimizers()

Collects learnable parameters and configures the optimizer and learning rate scheduler. :returns: two lists containing the optimizer and the scheduler. :rtype: Tuple[List, List]

configure_sharded_model() → None

Hook to create modules in a distributed aware context. This is useful for when using sharded plugins, where we’d like to shard the model instantly, which is useful for extremely large models which can save memory and initialization time.

This hook is called during each of fit/val/test/predict stages in the same process, so ensure that implementation of this hook is idempotent.

cpu() → Self

See torch.nn.Module.cpu().

cuda(device: Optional[Union[device, int]] = None) → Self

Moves all model parameters and buffers to the GPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized.

Parameters

device – If specified, all parameters will be copied to that device. If None, the current CUDA device index will be used.

Returns

self

Return type

Module

property current_epoch: int

The current epoch in the Trainer, or 0 if not attached.

double() → Self

See torch.nn.Module.double().

eval() → T

Sets the module in evaluation mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False).

See Locally disabling gradient computation for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns

self

Return type

Module

property example_input_array: Optional[Union[Tensor, Tuple, Dict]]

The example input array is a specification of what the module can consume in the forward() method. The return type is interpreted as follows:

• Single tensor: It is assumed the model takes a single argument, i.e., model.forward(model.example_input_array)

• Tuple: The input array should be interpreted as a sequence of positional arguments, i.e., model.forward(*model.example_input_array)

• Dict: The input array represents named keyword arguments, i.e., model.forward(**model.example_input_array)

extra_repr() → str

Set the extra representation of the module

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

abstract fit(views: Iterable[ndarray], y=None, **kwargs)

Fits the model to the given data

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• y (None) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

self

Return type

object

fit_transform(views: Iterable[ndarray], **kwargs)

Fits the model to the given data and returns the transformed views

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

transformed_views

Return type

list of numpy arrays

float() → Self

See torch.nn.Module.float().

freeze() → None

Freeze all params for inference.

Example:

model = MyLightningModule(...)
model.freeze()

get_buffer(target: str) → Tensor

Returns the buffer given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The buffer referenced by target

Return type

torch.Tensor

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state() → Any

Returns any extra state to include in the module’s state_dict. Implement this and a corresponding set_extra_state() for your module if you need to store extra state. This function is called when building the module’s state_dict().

Note that extra state should be pickleable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns

Any extra state to store in the module’s state_dict

Return type

object

get_factor_loadings(views: Iterable[ndarray], normalize=True, **kwargs)

Returns the factor loadings for each view

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• normalize (bool, optional) – Whether to normalize the factor loadings. Default is True.

• kwargs (any additional keyword arguments required by the given model) –

Returns

factor_loadings

Return type

list of numpy arrays

get_parameter(target: str) → Parameter

Returns the parameter given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The Parameter referenced by target

Return type

torch.nn.Parameter

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_params(deep=True)

Get parameters for this estimator.

Parameters

deep (bool, default=True) – If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns

params – Parameter names mapped to their values.

Return type

dict

get_submodule(target: str) → Module

Returns the submodule given by target if it exists, otherwise throws an error.

For example, let’s say you have an nn.Module A that looks like this:

A(
    (net_b): Module(
        (net_c): Module(
            (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2))
        )
        (linear): Linear(in_features=100, out_features=200, bias=True)
    )
)

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Parameters

target – The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns

The submodule referenced by target

Return type

torch.nn.Module

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Module

property global_rank: int

The index of the current process across all nodes and devices.

property global_step: int

Total training batches seen across all epochs.

If no Trainer is attached, this propery is 0.

half() → Self

See torch.nn.Module.half().

property hparams: Union[AttributeDict, MutableMapping]

The collection of hyperparameters saved with save_hyperparameters(). It is mutable by the user. For the frozen set of initial hyperparameters, use hparams_initial.

Returns

Mutable hyperparameters dictionary

property hparams_initial: AttributeDict

The collection of hyperparameters saved with save_hyperparameters(). These contents are read-only. Manual updates to the saved hyperparameters can instead be performed through hparams.

Returns

immutable initial hyperparameters

Return type

AttributeDict

ipu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the IPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on IPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

classmethod load_from_checkpoint(checkpoint_path: Union[str, Path, IO], map_location: Optional[Union[device, str, int, Callable[[Union[device, str, int]], Union[device, str, int]], Dict[Union[device, str, int], Union[device, str, int]]]] = None, hparams_file: Optional[Union[str, Path]] = None, strict: bool = True, **kwargs: Any) → Self

Primary way of loading a model from a checkpoint. When Lightning saves a checkpoint it stores the arguments passed to __init__ in the checkpoint under "hyper_parameters".

Any arguments specified through **kwargs will override args stored in "hyper_parameters".

Parameters

• checkpoint_path – Path to checkpoint. This can also be a URL, or file-like object

• map_location – If your checkpoint saved a GPU model and you now load on CPUs or a different number of GPUs, use this to map to the new setup. The behaviour is the same as in torch.load().

• hparams_file –

  Optional path to a .yaml or .csv file with hierarchical structure as in this example:

  drop_prob: 0.2
  dataloader:
      batch_size: 32
  

  You most likely won’t need this since Lightning will always save the hyperparameters to the checkpoint. However, if your checkpoint weights don’t have the hyperparameters saved, use this method to pass in a .yaml file with the hparams you’d like to use. These will be converted into a dict and passed into your LightningModule for use.

  If your model’s hparams argument is Namespace and .yaml file has hierarchical structure, you need to refactor your model to treat hparams as dict.

• strict – Whether to strictly enforce that the keys in checkpoint_path match the keys returned by this module’s state dict.

• **kwargs – Any extra keyword args needed to init the model. Can also be used to override saved hyperparameter values.

Returns

LightningModule instance with loaded weights and hyperparameters (if available).

Note

load_from_checkpoint is a class method. You should use your LightningModule class to call it instead of the LightningModule instance.

Example:

# load weights without mapping ...
model = MyLightningModule.load_from_checkpoint('path/to/checkpoint.ckpt')

# or load weights mapping all weights from GPU 1 to GPU 0 ...
map_location = {'cuda:1':'cuda:0'}
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    map_location=map_location
)

# or load weights and hyperparameters from separate files.
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    hparams_file='/path/to/hparams_file.yaml'
)

# override some of the params with new values
model = MyLightningModule.load_from_checkpoint(
    PATH,
    num_layers=128,
    pretrained_ckpt_path=NEW_PATH,
)

# predict
pretrained_model.eval()
pretrained_model.freeze()
y_hat = pretrained_model(x)

load_state_dict(state_dict: Mapping[str, Any], strict: bool = True)

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Parameters

• state_dict (dict) – a dict containing parameters and persistent buffers.

• strict (bool, optional) – whether to strictly enforce that the keys in state_dict match the keys returned by this module’s state_dict() function. Default: True

Returns

• missing_keys is a list of str containing the missing keys

• unexpected_keys is a list of str containing the unexpected keys

Return type

NamedTuple with missing_keys and unexpected_keys fields

Note

If a parameter or buffer is registered as None and its corresponding key exists in state_dict, load_state_dict() will raise a RuntimeError.

property local_rank: int

The index of the current process within a single node.

log(name: str, value: Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, metric_attribute: Optional[str] = None, rank_zero_only: bool = False) → None

Log a key, value pair.

Example:

self.log('train_loss', loss)

The default behavior per hook is documented here: extensions/logging:Automatic Logging.

Parameters

• name – key to log.

• value – value to log. Can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress bar.

• logger – if True logs to the logger.

• on_step – if True logs at this step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto detach the graph.

• sync_dist – if True, reduces the metric across devices. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the DDP group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple dataloaders). If False, user needs to give unique names for each dataloader to not mix the values.

• batch_size – Current batch_size. This will be directly inferred from the loaded batch, but for some data structures you might need to explicitly provide it.

• metric_attribute – To restore the metric state, Lightning requires the reference of the torchmetrics.Metric in your model. This is found automatically if it is a model attribute.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_dict(dictionary: Mapping[str, Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, rank_zero_only: bool = False) → None

Log a dictionary of values at once.

Example:

values = {'loss': loss, 'acc': acc, ..., 'metric_n': metric_n}
self.log_dict(values)

Parameters

• dictionary – key value pairs. The values can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress base.

• logger – if True logs to the logger.

• on_step – if True logs at this step. None auto-logs for training_step but not validation/test_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. None auto-logs for val/test step but not training_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto-detach the graph

• sync_dist – if True, reduces the metric across GPUs/TPUs. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the ddp group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple). If False, user needs to give unique names for each dataloader to not mix values.

• batch_size – Current batch size. This will be directly inferred from the loaded batch, but some data structures might need to explicitly provide it.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_grad_norm(grad_norm_dict: Dict[str, float]) → None

Override this method to change the default behaviour of log_grad_norm.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

grad_norm_dict – Dictionary containing current grad norm metrics

Example:

# DEFAULT
def log_grad_norm(self, grad_norm_dict):
    self.log_dict(grad_norm_dict, on_step=True, on_epoch=True, prog_bar=False, logger=True)

property logger: Optional[Logger]

Reference to the logger object in the Trainer.

property loggers: List[Logger]

Reference to the list of loggers in the Trainer.

lr_scheduler_step(scheduler: Union[_LRScheduler, ReduceLROnPlateau], optimizer_idx: int, metric: Optional[Any]) → None

Override this method to adjust the default way the Trainer calls each scheduler. By default, Lightning calls step() and as shown in the example for each scheduler based on its interval.

Parameters

• scheduler – Learning rate scheduler.

• optimizer_idx – Index of the optimizer associated with this scheduler.

• metric – Value of the monitor used for schedulers like ReduceLROnPlateau.

Examples:

# DEFAULT
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    if metric is None:
        scheduler.step()
    else:
        scheduler.step(metric)

# Alternative way to update schedulers if it requires an epoch value
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    scheduler.step(epoch=self.current_epoch)

lr_schedulers() → Union[None, List[Union[_LRScheduler, ReduceLROnPlateau]], _LRScheduler, ReduceLROnPlateau]

Returns the learning rate scheduler(s) that are being used during training. Useful for manual optimization.

Returns

A single scheduler, or a list of schedulers in case multiple ones are present, or None if no schedulers were returned in configure_optimizers().

manual_backward(loss: Tensor, *args: Any, **kwargs: Any) → None

Call this directly from your training_step() when doing optimizations manually. By using this, Lightning can ensure that all the proper scaling gets applied when using mixed precision.

See manual optimization for more examples.

Example:

def training_step(...):
    opt = self.optimizers()
    loss = ...
    opt.zero_grad()
    # automatically applies scaling, etc...
    self.manual_backward(loss)
    opt.step()

Parameters

• loss – The tensor on which to compute gradients. Must have a graph attached.

• *args – Additional positional arguments to be forwarded to backward()

• **kwargs – Additional keyword arguments to be forwarded to backward()

modules() → Iterator[Module]

Returns an iterator over all modules in the network.

Yields

Module – a module in the network

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.modules()):
...     print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Tensor]]

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Parameters

• prefix (str) – prefix to prepend to all buffer names.

• recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

(str, torch.Tensor) – Tuple containing the name and buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, buf in self.named_buffers():
>>>    if name in ['running_var']:
>>>        print(buf.size())

named_children() → Iterator[Tuple[str, Module]]

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields

(str, Module) – Tuple containing a name and child module

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, module in model.named_children():
>>>     if name in ['conv4', 'conv5']:
>>>         print(module)

named_modules(memo: Optional[Set[Module]] = None, prefix: str = '', remove_duplicate: bool = True)

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Parameters

• memo – a memo to store the set of modules already added to the result

• prefix – a prefix that will be added to the name of the module

• remove_duplicate – whether to remove the duplicated module instances in the result or not

Yields

(str, Module) – Tuple of name and module

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.named_modules()):
...     print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Parameter]]

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Parameters

• prefix (str) – prefix to prepend to all parameter names.

• recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

(str, Parameter) – Tuple containing the name and parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, param in self.named_parameters():
>>>    if name in ['bias']:
>>>        print(param.size())

on_after_backward() → None

Called after loss.backward() and before optimizers are stepped.

Note

If using native AMP, the gradients will not be unscaled at this point. Use the on_before_optimizer_step if you need the unscaled gradients.

on_after_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch after it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_after_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = gpu_transforms(batch['x'])
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• on_before_batch_transfer()

• transfer_batch_to_device()

on_before_backward(loss: Tensor) → None

Called before loss.backward().

Parameters

loss – Loss divided by number of batches for gradient accumulation and scaled if using native AMP.

on_before_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch before it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_before_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = transforms(batch['x'])
    return batch

See also

• on_after_batch_transfer()

• transfer_batch_to_device()

on_before_optimizer_step(optimizer: Optimizer, optimizer_idx: int) → None

Called before optimizer.step().

If using gradient accumulation, the hook is called once the gradients have been accumulated. See: :paramref:`~pytorch_lightning.trainer.Trainer.accumulate_grad_batches`.

If using native AMP, the loss will be unscaled before calling this hook. See these docs for more information on the scaling of gradients.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

Example:

def on_before_optimizer_step(self, optimizer, optimizer_idx):
    # example to inspect gradient information in tensorboard
    if self.trainer.global_step % 25 == 0:  # don't make the tf file huge
        for k, v in self.named_parameters():
            self.logger.experiment.add_histogram(
                tag=k, values=v.grad, global_step=self.trainer.global_step
            )

on_before_zero_grad(optimizer: Optimizer) → None

Called after training_step() and before optimizer.zero_grad().

Called in the training loop after taking an optimizer step and before zeroing grads. Good place to inspect weight information with weights updated.

This is where it is called:

for optimizer in optimizers:
    out = training_step(...)

    model.on_before_zero_grad(optimizer) # < ---- called here
    optimizer.zero_grad()

    backward()

Parameters

optimizer – The optimizer for which grads should be zeroed.

on_fit_end() → None

Called at the very end of fit.

If on DDP it is called on every process

on_fit_start() → None

Called at the very beginning of fit.

If on DDP it is called on every process

property on_gpu: bool

Returns True if this model is currently located on a GPU.

Useful to set flags around the LightningModule for different CPU vs GPU behavior.

on_load_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning to restore your model. If you saved something with on_save_checkpoint() this is your chance to restore this.

Parameters

checkpoint – Loaded checkpoint

Example:

def on_load_checkpoint(self, checkpoint):
    # 99% of the time you don't need to implement this method
    self.something_cool_i_want_to_save = checkpoint['something_cool_i_want_to_save']

Note

Lightning auto-restores global step, epoch, and train state including amp scaling. There is no need for you to restore anything regarding training.

on_predict_batch_end(outputs: Optional[Any], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop after the batch.

Parameters

• outputs – The outputs of predict_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_end() → None

Called at the end of predicting.

on_predict_epoch_end(results: List[Any]) → None

Called at the end of predicting.

on_predict_epoch_start() → None

Called at the beginning of predicting.

on_predict_model_eval() → None

Sets the model to eval during the predict loop.

on_predict_start() → None

Called at the beginning of predicting.

on_save_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning when saving a checkpoint to give you a chance to store anything else you might want to save.

Parameters

checkpoint – The full checkpoint dictionary before it gets dumped to a file. Implementations of this hook can insert additional data into this dictionary.

Example:

def on_save_checkpoint(self, checkpoint):
    # 99% of use cases you don't need to implement this method
    checkpoint['something_cool_i_want_to_save'] = my_cool_pickable_object

Note

Lightning saves all aspects of training (epoch, global step, etc…) including amp scaling. There is no need for you to store anything about training.

on_test_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop after the batch.

Parameters

• outputs – The outputs of test_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_end() → None

Called at the end of testing.

on_test_epoch_end() → None

Called in the test loop at the very end of the epoch.

on_test_epoch_start() → None

Called in the test loop at the very beginning of the epoch.

on_test_model_eval() → None

Sets the model to eval during the test loop.

on_test_model_train() → None

Sets the model to train during the test loop.

on_test_start() → None

Called at the beginning of testing.

on_train_batch_end(outputs: Union[Tensor, Dict[str, Any]], batch: Any, batch_idx: int) → None

Called in the training loop after the batch.

Parameters

• outputs – The outputs of training_step_end(training_step(x))

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_batch_start(batch: Any, batch_idx: int) → Optional[int]

Called in the training loop before anything happens for that batch.

If you return -1 here, you will skip training for the rest of the current epoch.

Parameters

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_end() → None

Called at the end of training before logger experiment is closed.

on_train_epoch_end() → None

Called in the training loop at the very end of the epoch.

To access all batch outputs at the end of the epoch, either:

1. Implement training_epoch_end in the LightningModule OR

2. Cache data across steps on the attribute(s) of the LightningModule and access them in this hook

on_train_epoch_start() → None

Called in the training loop at the very beginning of the epoch.

on_train_start() → None

Called at the beginning of training after sanity check.

on_validation_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop after the batch.

Parameters

• outputs – The outputs of validation_step_end(validation_step(x))

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_end() → None

Called at the end of validation.

on_validation_epoch_end() → None

Called in the validation loop at the very end of the epoch.

on_validation_epoch_start() → None

Called in the validation loop at the very beginning of the epoch.

on_validation_model_eval() → None

Sets the model to eval during the val loop.

on_validation_model_train() → None

Sets the model to train during the val loop.

on_validation_start() → None

Called at the beginning of validation.

optimizer_step(epoch: int, batch_idx: int, optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int = 0, optimizer_closure: Optional[Callable[[], Any]] = None, on_tpu: bool = False, using_native_amp: bool = False, using_lbfgs: bool = False) → None

Override this method to adjust the default way the Trainer calls each optimizer.

By default, Lightning calls step() and zero_grad() as shown in the example once per optimizer. This method (and zero_grad()) won’t be called during the accumulation phase when Trainer(accumulate_grad_batches != 1). Overriding this hook has no benefit with manual optimization.

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers, this indexes into that list.

• optimizer_closure – The optimizer closure. This closure must be executed as it includes the calls to training_step(), optimizer.zero_grad(), and backward().

• on_tpu – True if TPU backward is required

• using_native_amp – True if using native amp

• using_lbfgs – True if the matching optimizer is torch.optim.LBFGS

Examples:

# DEFAULT
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    optimizer.step(closure=optimizer_closure)

# Alternating schedule for optimizer steps (i.e.: GANs)
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    # update generator opt every step
    if optimizer_idx == 0:
        optimizer.step(closure=optimizer_closure)

    # update discriminator opt every 2 steps
    if optimizer_idx == 1:
        if (batch_idx + 1) % 2 == 0 :
            optimizer.step(closure=optimizer_closure)
        else:
            # call the closure by itself to run `training_step` + `backward` without an optimizer step
            optimizer_closure()

    # ...
    # add as many optimizers as you want

Here’s another example showing how to use this for more advanced things such as learning rate warm-up:

# learning rate warm-up
def optimizer_step(
    self,
    epoch,
    batch_idx,
    optimizer,
    optimizer_idx,
    optimizer_closure,
    on_tpu,
    using_native_amp,
    using_lbfgs,
):
    # update params
    optimizer.step(closure=optimizer_closure)

    # manually warm up lr without a scheduler
    if self.trainer.global_step < 500:
        lr_scale = min(1.0, float(self.trainer.global_step + 1) / 500.0)
        for pg in optimizer.param_groups:
            pg["lr"] = lr_scale * self.learning_rate

optimizer_zero_grad(epoch: int, batch_idx: int, optimizer: Optimizer, optimizer_idx: int) → None

Override this method to change the default behaviour of optimizer.zero_grad().

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers this indexes into that list.

Examples:

# DEFAULT
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad()

# Set gradients to `None` instead of zero to improve performance.
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad(set_to_none=True)

See torch.optim.Optimizer.zero_grad() for the explanation of the above example.

optimizers(use_pl_optimizer: bool = True) → Union[Optimizer, LightningOptimizer, List[Optimizer], List[LightningOptimizer]]

Returns the optimizer(s) that are being used during training. Useful for manual optimization.

Parameters

use_pl_optimizer – If True, will wrap the optimizer(s) in a LightningOptimizer for automatic handling of precision and profiling.

Returns

A single optimizer, or a list of optimizers in case multiple ones are present.

parameters(recurse: bool = True) → Iterator[Parameter]

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Parameters

recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

Parameter – module parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for param in model.parameters():
>>>     print(type(param), param.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

predict_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for prediction.

It’s recommended that all data downloads and preparation happen in prepare_data().

• predict()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying prediction samples.

Note

In the case where you return multiple prediction dataloaders, the predict_step() will have an argument dataloader_idx which matches the order here.

predict_step(batch: Any, batch_idx: int, dataloader_idx: int = 0) → Any

Step function called during predict(). By default, it calls forward(). Override to add any processing logic.

The predict_step() is used to scale inference on multi-devices.

To prevent an OOM error, it is possible to use BasePredictionWriter callback to write the predictions to disk or database after each batch or on epoch end.

The BasePredictionWriter should be used while using a spawn based accelerator. This happens for Trainer(strategy="ddp_spawn") or training on 8 TPU cores with Trainer(accelerator="tpu", devices=8) as predictions won’t be returned.

Example

class MyModel(LightningModule):

    def predict_step(self, batch, batch_idx, dataloader_idx=0):
        return self(batch)

dm = ...
model = MyModel()
trainer = Trainer(accelerator="gpu", devices=2)
predictions = trainer.predict(model, dm)

Parameters

• batch – Current batch.

• batch_idx – Index of current batch.

• dataloader_idx – Index of the current dataloader.

Returns

Predicted output

prepare_data() → None

Use this to download and prepare data. Downloading and saving data with multiple processes (distributed settings) will result in corrupted data. Lightning ensures this method is called only within a single process, so you can safely add your downloading logic within.

Warning

DO NOT set state to the model (use setup instead) since this is NOT called on every device

Example:

def prepare_data(self):
    # good
    download_data()
    tokenize()
    etc()

    # bad
    self.split = data_split
    self.some_state = some_other_state()

In a distributed environment, prepare_data can be called in two ways (using prepare_data_per_node)

1. Once per node. This is the default and is only called on LOCAL_RANK=0.

2. Once in total. Only called on GLOBAL_RANK=0.

Example:

# DEFAULT
# called once per node on LOCAL_RANK=0 of that node
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = True


# call on GLOBAL_RANK=0 (great for shared file systems)
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = False

This is called before requesting the dataloaders:

model.prepare_data()
initialize_distributed()
model.setup(stage)
model.train_dataloader()
model.val_dataloader()
model.test_dataloader()
model.predict_dataloader()

print(*args: Any, **kwargs: Any) → None

Prints only from process 0. Use this in any distributed mode to log only once.

Parameters

• *args – The thing to print. The same as for Python’s built-in print function.

• **kwargs – The same as for Python’s built-in print function.

Example:

def forward(self, x):
    self.print(x, 'in forward')

register_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

This function is deprecated in favor of register_full_backward_hook() and the behavior of this function will change in future versions.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_buffer(name: str, tensor: Optional[Tensor], persistent: bool = True) → None

Adds a buffer to the module.

This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the module’s state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module’s state_dict.

Buffers can be accessed as attributes using given names.

Parameters

• name (str) – name of the buffer. The buffer can be accessed from this module using the given name

• tensor (Tensor or None) – buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module’s state_dict.

• persistent (bool) – whether the buffer is part of this module’s state_dict.

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> self.register_buffer('running_mean', torch.zeros(num_features))

register_forward_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output. It should have the following signature:

hook(module, input, output) -> None or modified output

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward() is called.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_forward_pre_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked. It should have the following signature:

hook(module, input) -> None or modified input

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned(unless that value is already a tuple).

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_full_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

The hook will be called every time the gradients with respect to module inputs are computed. The hook should have the following signature:

hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module’s forward function.

Warning

Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_load_state_dict_post_hook(hook)

Registers a post hook to be run after module’s load_state_dict is called.

It should have the following signature::

hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict() with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearning out both missing and unexpected keys will avoid an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_module(name: str, module: Optional[Module]) → None

Alias for add_module().

register_parameter(name: str, param: Optional[Parameter]) → None

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Parameters

• name (str) – name of the parameter. The parameter can be accessed from this module using the given name

• param (Parameter or None) – parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module’s state_dict.

requires_grad_(requires_grad: bool = True) → T

Change if autograd should record operations on parameters in this module.

This method sets the parameters’ requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See Locally disabling gradient computation for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Parameters

requires_grad (bool) – whether autograd should record operations on parameters in this module. Default: True.

Returns

self

Return type

Module

save_hyperparameters(*args: Any, ignore: Optional[Union[Sequence[str], str]] = None, frame: Optional[frame] = None, logger: bool = True) → None

Save arguments to hparams attribute.

Parameters

• args – single object of dict, NameSpace or OmegaConf or string names or arguments from class __init__

• ignore – an argument name or a list of argument names from class __init__ to be ignored

• frame – a frame object. Default is None

• logger – Whether to send the hyperparameters to the logger. Default: True

Example::

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # manually assign arguments
...         self.save_hyperparameters('arg1', 'arg3')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

>>> class AutomaticArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # equivalent automatic
...         self.save_hyperparameters()
...     def forward(self, *args, **kwargs):
...         ...
>>> model = AutomaticArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg2": abc
"arg3": 3.14

>>> class SingleArgModel(HyperparametersMixin):
...     def __init__(self, params):
...         super().__init__()
...         # manually assign single argument
...         self.save_hyperparameters(params)
...     def forward(self, *args, **kwargs):
...         ...
>>> model = SingleArgModel(Namespace(p1=1, p2='abc', p3=3.14))
>>> model.hparams
"p1": 1
"p2": abc
"p3": 3.14

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # pass argument(s) to ignore as a string or in a list
...         self.save_hyperparameters(ignore='arg2')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

set_extra_state(state: Any)

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Parameters

state (dict) – Extra state from the state_dict

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters

**params (dict) – Estimator parameters.

Returns

self – Estimator instance.

Return type

estimator instance

setup(stage: str) → None

Called at the beginning of fit (train + validate), validate, test, or predict. This is a good hook when you need to build models dynamically or adjust something about them. This hook is called on every process when using DDP.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

Example:

class LitModel(...):
    def __init__(self):
        self.l1 = None

    def prepare_data(self):
        download_data()
        tokenize()

        # don't do this
        self.something = else

    def setup(self, stage):
        data = load_data(...)
        self.l1 = nn.Linear(28, data.num_classes)

share_memory() → T

See torch.Tensor.share_memory_()

state_dict(*args, destination=None, prefix='', keep_vars=False)

Returns a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

Note

The returned object is a shallow copy. It contains references to the module’s parameters and buffers.

Warning

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

Warning

Please avoid the use of argument destination as it is not designed for end-users.

Parameters

• destination (dict, optional) – If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned. Default: None.

• prefix (str, optional) – a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.

• keep_vars (bool, optional) – by default the Tensor s returned in the state dict are detached from autograd. If it’s set to True, detaching will not be performed. Default: False.

Returns

a dictionary containing a whole state of the module

Return type

dict

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> module.state_dict().keys()
['bias', 'weight']

tbptt_split_batch(batch: Any, split_size: int) → List[Any]

When using truncated backpropagation through time, each batch must be split along the time dimension. Lightning handles this by default, but for custom behavior override this function.

Parameters

• batch – Current batch

• split_size – The size of the split

Returns

List of batch splits. Each split will be passed to training_step() to enable truncated back propagation through time. The default implementation splits root level Tensors and Sequences at dim=1 (i.e. time dim). It assumes that each time dim is the same length.

Examples:

def tbptt_split_batch(self, batch, split_size):
    splits = []
    for t in range(0, time_dims[0], split_size):
        batch_split = []
        for i, x in enumerate(batch):
            if isinstance(x, torch.Tensor):
                split_x = x[:, t:t + split_size]
            elif isinstance(x, collections.abc.Sequence):
                split_x = [None] * len(x)
                for batch_idx in range(len(x)):
                  split_x[batch_idx] = x[batch_idx][t:t + split_size]
            batch_split.append(split_x)
        splits.append(batch_split)
    return splits

Note

Called in the training loop after on_train_batch_start() if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0. Each returned batch split is passed separately to training_step().

teardown(stage: str) → None

Called at the end of fit (train + validate), validate, test, or predict.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

test_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for testing.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• test()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying testing samples.

Example:

def test_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def test_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a test dataset and a test_step(), you don’t need to implement this method.

Note

In the case where you return multiple test dataloaders, the test_step() will have an argument dataloader_idx which matches the order here.

test_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of a test epoch with the output of all test steps.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

outputs – List of outputs you defined in test_step_end(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader

Returns

None

Note

If you didn’t define a test_step(), this won’t be called.

Examples

With a single dataloader:

def test_epoch_end(self, outputs):
    # do something with the outputs of all test batches
    all_test_preds = test_step_outputs.predictions

    some_result = calc_all_results(all_test_preds)
    self.log(some_result)

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each test step for that dataloader.

def test_epoch_end(self, outputs):
    final_value = 0
    for dataloader_outputs in outputs:
        for test_step_out in dataloader_outputs:
            # do something
            final_value += test_step_out

    self.log("final_metric", final_value)

test_step(batch, batch_idx)

Operates on a single batch of data from the test set. In this step you’d normally generate examples or calculate anything of interest such as accuracy.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_id – The index of the dataloader that produced this batch. (only if multiple test dataloaders used).

Returns

Any of.

• Any object or value

• None - Testing will skip to the next batch

# if you have one test dataloader:
def test_step(self, batch, batch_idx):
    ...


# if you have multiple test dataloaders:
def test_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single test dataset
def test_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    test_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'test_loss': loss, 'test_acc': test_acc})

If you pass in multiple test dataloaders, test_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple test dataloaders
def test_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to test you don’t need to implement this method.

Note

When the test_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of the test epoch, the model goes back to training mode and gradients are enabled.

test_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when testing with DP because test_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [test_step(sub_batch) for sub_batch in sub_batches]
test_step_end(step_output)

Parameters

step_output – What you return in test_step() for each batch part.

Returns

None or anything

# WITHOUT test_step_end
# if used in DP, this batch is 1/num_gpus large
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    loss = self.softmax(out)
    self.log("test_loss", loss)


# --------------
# with test_step_end to do softmax over the full batch
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return out


def test_step_end(self, output_results):
    # this out is now the full size of the batch
    all_test_step_outs = output_results.out
    loss = nce_loss(all_test_step_outs)
    self.log("test_loss", loss)

See also

See the Multi GPU Training guide for more details.

to(*args: Any, **kwargs: Any) → Self

See torch.nn.Module.to().

to_empty(*, device: Union[str, device]) → T

Moves the parameters and buffers to the specified device without copying storage.

Parameters

device (torch.device) – The desired device of the parameters and buffers in this module.

Returns

self

Return type

Module

to_onnx(file_path: Union[str, Path], input_sample: Optional[Any] = None, **kwargs: Any) → None

Saves the model in ONNX format.

Parameters

• file_path – The path of the file the onnx model should be saved to.

• input_sample – An input for tracing. Default: None (Use self.example_input_array)

• **kwargs – Will be passed to torch.onnx.export function.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))

>>> with tempfile.NamedTemporaryFile(suffix='.onnx', delete=False) as tmpfile:
...     model = SimpleModel()
...     input_sample = torch.randn((1, 64))
...     model.to_onnx(tmpfile.name, input_sample, export_params=True)
...     os.path.isfile(tmpfile.name)
True

to_torchscript(file_path: Optional[Union[str, Path]] = None, method: Optional[str] = 'script', example_inputs: Optional[Any] = None, **kwargs: Any) → Union[ScriptModule, Dict[str, ScriptModule]]

By default compiles the whole model to a ScriptModule. If you want to use tracing, please provided the argument method='trace' and make sure that either the example_inputs argument is provided, or the model has example_input_array set. If you would like to customize the modules that are scripted you should override this method. In case you want to return multiple modules, we recommend using a dictionary.

Parameters

• file_path – Path where to save the torchscript. Default: None (no file saved).

• method – Whether to use TorchScript’s script or trace method. Default: ‘script’

• example_inputs – An input to be used to do tracing when method is set to ‘trace’. Default: None (uses example_input_array)

• **kwargs – Additional arguments that will be passed to the torch.jit.script() or torch.jit.trace() function.

Note

• Requires the implementation of the forward() method.

• The exported script will be set to evaluation mode.

• It is recommended that you install the latest supported version of PyTorch to use this feature without limitations. See also the torch.jit documentation for supported features.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))
...
>>> model = SimpleModel()
>>> model.to_torchscript(file_path="model.pt")  
>>> os.path.isfile("model.pt")  
>>> torch.jit.save(model.to_torchscript(file_path="model_trace.pt", method='trace', 
...                                     example_inputs=torch.randn(1, 64)))  
>>> os.path.isfile("model_trace.pt")  
True

Returns

This LightningModule as a torchscript, regardless of whether file_path is defined or not.

toggle_optimizer(optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int) → None

Makes sure only the gradients of the current optimizer’s parameters are calculated in the training step to prevent dangling gradients in multiple-optimizer setup.

This is only called automatically when automatic optimization is enabled and multiple optimizers are used. It works with untoggle_optimizer() to make sure param_requires_grad_state is properly reset.

Parameters

• optimizer – The optimizer to toggle.

• optimizer_idx – The index of the optimizer to toggle.

train(mode: bool = True) → T

Sets the module in training mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

Parameters

mode (bool) – whether to set training mode (True) or evaluation mode (False). Default: True.

Returns

self

Return type

Module

train_dataloader() → Union[DataLoader, Sequence[DataLoader], Sequence[Sequence[DataLoader]], Sequence[Dict[str, DataLoader]], Dict[str, DataLoader], Dict[str, Dict[str, DataLoader]], Dict[str, Sequence[DataLoader]]]

Implement one or more PyTorch DataLoaders for training.

Returns

A collection of torch.utils.data.DataLoader specifying training samples. In the case of multiple dataloaders, please see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• fit()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Example:

# single dataloader
def train_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=True, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=True
    )
    return loader

# multiple dataloaders, return as list
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a list of tensors: [batch_mnist, batch_cifar]
    return [mnist_loader, cifar_loader]

# multiple dataloader, return as dict
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a dict of tensors: {'mnist': batch_mnist, 'cifar': batch_cifar}
    return {'mnist': mnist_loader, 'cifar': cifar_loader}

training_epoch_end(outputs: List[Union[Tensor, Dict[str, Any]]]) → None

Called at the end of the training epoch with the outputs of all training steps. Use this in case you need to do something with all the outputs returned by training_step().

# the pseudocode for these calls
train_outs = []
for train_batch in train_data:
    out = training_step(train_batch)
    train_outs.append(out)
training_epoch_end(train_outs)

Parameters

outputs – List of outputs you defined in training_step(). If there are multiple optimizers or when using truncated_bptt_steps > 0, the lists have the dimensions (n_batches, tbptt_steps, n_optimizers). Dimensions of length 1 are squeezed.

Returns

None

Note

If this method is not overridden, this won’t be called.

def training_epoch_end(self, training_step_outputs):
    # do something with all training_step outputs
    for out in training_step_outputs:
        ...

training_step(batch, batch_idx)

Here you compute and return the training loss and some additional metrics for e.g. the progress bar or logger.

Parameters

• batch (Tensor | (Tensor, …) | [Tensor, …]) – The output of your DataLoader. A tensor, tuple or list.

• batch_idx (int) – Integer displaying index of this batch

• optimizer_idx (int) – When using multiple optimizers, this argument will also be present.

• hiddens (Any) – Passed in if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0.

Returns

Any of.

• Tensor - The loss tensor

• dict - A dictionary. Can include any keys, but must include the key 'loss'

• None - Training will skip to the next batch. This is only for automatic optimization.

  This is not supported for multi-GPU, TPU, IPU, or DeepSpeed.

In this step you’d normally do the forward pass and calculate the loss for a batch. You can also do fancier things like multiple forward passes or something model specific.

Example:

def training_step(self, batch, batch_idx):
    x, y, z = batch
    out = self.encoder(x)
    loss = self.loss(out, x)
    return loss

If you define multiple optimizers, this step will be called with an additional optimizer_idx parameter.

# Multiple optimizers (e.g.: GANs)
def training_step(self, batch, batch_idx, optimizer_idx):
    if optimizer_idx == 0:
        # do training_step with encoder
        ...
    if optimizer_idx == 1:
        # do training_step with decoder
        ...

If you add truncated back propagation through time you will also get an additional argument with the hidden states of the previous step.

# Truncated back-propagation through time
def training_step(self, batch, batch_idx, hiddens):
    # hiddens are the hidden states from the previous truncated backprop step
    out, hiddens = self.lstm(data, hiddens)
    loss = ...
    return {"loss": loss, "hiddens": hiddens}

Note

The loss value shown in the progress bar is smoothed (averaged) over the last values, so it differs from the actual loss returned in train/validation step.

Note

When accumulate_grad_batches > 1, the loss returned here will be automatically normalized by accumulate_grad_batches internally.

training_step_end(step_output: Union[Tensor, Dict[str, Any]]) → Union[Tensor, Dict[str, Any]]

Use this when training with dp because training_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [training_step(sub_batch) for sub_batch in sub_batches]
training_step_end(step_output)

Parameters

step_output – What you return in training_step for each batch part.

Returns

Anything

When using the DP strategy, only a portion of the batch is inside the training_step:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)

    # softmax uses only a portion of the batch in the denominator
    loss = self.softmax(out)
    loss = nce_loss(loss)
    return loss

If you wish to do something with all the parts of the batch, then use this method to do it:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return {"pred": out}


def training_step_end(self, training_step_outputs):
    gpu_0_pred = training_step_outputs[0]["pred"]
    gpu_1_pred = training_step_outputs[1]["pred"]
    gpu_n_pred = training_step_outputs[n]["pred"]

    # this softmax now uses the full batch
    loss = nce_loss([gpu_0_pred, gpu_1_pred, gpu_n_pred])
    return loss

See also

See the Multi GPU Training guide for more details.

transfer_batch_to_device(batch: Any, device: device, dataloader_idx: int) → Any

Override this hook if your DataLoader returns tensors wrapped in a custom data structure.

The data types listed below (and any arbitrary nesting of them) are supported out of the box:

• torch.Tensor or anything that implements .to(…)

• list

• dict

• tuple

For anything else, you need to define how the data is moved to the target device (CPU, GPU, TPU, …).

Note

This hook should only transfer the data and not modify it, nor should it move the data to any other device than the one passed in as argument (unless you know what you are doing). To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be transferred to a new device.

• device – The target device as defined in PyTorch.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A reference to the data on the new device.

Example:

def transfer_batch_to_device(self, batch, device, dataloader_idx):
    if isinstance(batch, CustomBatch):
        # move all tensors in your custom data structure to the device
        batch.samples = batch.samples.to(device)
        batch.targets = batch.targets.to(device)
    elif dataloader_idx == 0:
        # skip device transfer for the first dataloader or anything you wish
        pass
    else:
        batch = super().transfer_batch_to_device(data, device, dataloader_idx)
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• move_data_to_device()

• apply_to_collection()

transform(loader: DataLoader)

Parameters

loader – a dataloader that matches the structure of that used for training

Returns

transformed views

property truncated_bptt_steps: int

Enables Truncated Backpropagation Through Time in the Trainer when set to a positive integer.

It represents the number of times training_step() gets called before backpropagation. If this is > 0, the training_step() receives an additional argument hiddens and is expected to return a hidden state.

type(dst_type: Union[str, dtype]) → Self

See torch.nn.Module.type().

unfreeze() → None

Unfreeze all parameters for training.

model = MyLightningModule(...)
model.unfreeze()

untoggle_optimizer(optimizer_idx: int) → None

Resets the state of required gradients that were toggled with toggle_optimizer().

This is only called automatically when automatic optimization is enabled and multiple optimizers are used.

Parameters

optimizer_idx – The index of the optimizer to untoggle.

val_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for validation.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

It’s recommended that all data downloads and preparation happen in prepare_data().

• fit()

• validate()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying validation samples.

Examples:

def val_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False,
                    transform=transform, download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def val_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a validation dataset and a validation_step(), you don’t need to implement this method.

Note

In the case where you return multiple validation dataloaders, the validation_step() will have an argument dataloader_idx which matches the order here.

validation_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of the validation epoch with the outputs of all validation steps.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

outputs – List of outputs you defined in validation_step(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader.

Returns

None

Note

If you didn’t define a validation_step(), this won’t be called.

Examples

With a single dataloader:

def validation_epoch_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each validation step for that dataloader.

def validation_epoch_end(self, outputs):
    for dataloader_output_result in outputs:
        dataloader_outs = dataloader_output_result.dataloader_i_outputs

    self.log("final_metric", final_value)

validation_step(batch, batch_idx)

Operates on a single batch of data from the validation set. In this step you’d might generate examples or calculate anything of interest like accuracy.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_idx – The index of the dataloader that produced this batch. (only if multiple val dataloaders used)

Returns

• Any object or value

• None - Validation will skip to the next batch

# pseudocode of order
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    if defined("validation_step_end"):
        out = validation_step_end(out)
    val_outs.append(out)
val_outs = validation_epoch_end(val_outs)

# if you have one val dataloader:
def validation_step(self, batch, batch_idx):
    ...


# if you have multiple val dataloaders:
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single validation dataset
def validation_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    val_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'val_loss': loss, 'val_acc': val_acc})

If you pass in multiple val dataloaders, validation_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple validation dataloaders
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to validate you don’t need to implement this method.

Note

When the validation_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of validation, the model goes back to training mode and gradients are enabled.

validation_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when validating with dp because validation_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [validation_step(sub_batch) for sub_batch in sub_batches]
validation_step_end(step_output)

Parameters

step_output – What you return in validation_step() for each batch part.

Returns

None or anything

# WITHOUT validation_step_end
# if used in DP, this batch is 1/num_gpus large
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    loss = self.softmax(out)
    loss = nce_loss(loss)
    self.log("val_loss", loss)


# --------------
# with validation_step_end to do softmax over the full batch
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    return out


def validation_step_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

See also

See the Multi GPU Training guide for more details.

xpu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the XPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

zero_grad(set_to_none: bool = False) → None

Sets gradients of all model parameters to zero. See similar function under torch.optim.Optimizer for more context.

Parameters

set_to_none (bool) – instead of setting to zero, set the grads to None. See torch.optim.Optimizer.zero_grad() for details.

class cca_zoo.deepmodels.DCCAE(latent_dims: int, objective=<class 'cca_zoo.deepmodels.objectives.MCCA'>, encoders=None, decoders=None, r: float = 0, eps: float = 1e-05, lam=0.5, latent_dropout=0, img_dim=None, recon_loss_type='mse', **kwargs)

Bases: DCCA, _GenerativeMixin

A class used to fit a DCCAE model.

References

Wang, Weiran, et al. “On deep multi-view representation learning.” International conference on machine learning. PMLR, 2015.

forward(views, **kwargs)

Forward method for the model. Outputs latent encoding for each view

Parameters

• views –

• kwargs –

Returns

loss(views, **kwargs)

Loss for model :param views: :param args: :param kwargs: :return:

configure_callbacks()

Configure model-specific callbacks. When the model gets attached, e.g., when .fit() or .test() gets called, the list or a callback returned here will be merged with the list of callbacks passed to the Trainer’s callbacks argument. If a callback returned here has the same type as one or several callbacks already present in the Trainer’s callbacks list, it will take priority and replace them. In addition, Lightning will make sure ModelCheckpoint callbacks run last.

Returns

A callback or a list of callbacks which will extend the list of callbacks in the Trainer.

Example:

def configure_callbacks(self):
    early_stop = EarlyStopping(monitor="val_acc", mode="max")
    checkpoint = ModelCheckpoint(monitor="val_loss")
    return [early_stop, checkpoint]

add_module(name: str, module: Optional[Module]) → None

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Parameters

• name (str) – name of the child module. The child module can be accessed from this module using the given name

• module (Module) – child module to be added to the module.

all_gather(data: Union[Tensor, Dict, List, Tuple], group: Optional[Any] = None, sync_grads: bool = False) → Union[Tensor, Dict, List, Tuple]

Allows users to call self.all_gather() from the LightningModule, thus making the all_gather operation accelerator agnostic. all_gather is a function provided by accelerators to gather a tensor from several distributed processes.

Parameters

• data – int, float, tensor of shape (batch, …), or a (possibly nested) collection thereof.

• group – the process group to gather results from. Defaults to all processes (world)

• sync_grads – flag that allows users to synchronize gradients for the all_gather operation

Returns

A tensor of shape (world_size, batch, …), or if the input was a collection the output will also be a collection with tensors of this shape.

apply(fn: Callable[[Module], None]) → T

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also torch.nn.init).

Parameters

fn (Module -> None) – function to be applied to each submodule

Returns

self

Return type

Module

Example:

>>> @torch.no_grad()
>>> def init_weights(m):
>>>     print(m)
>>>     if type(m) == nn.Linear:
>>>         m.weight.fill_(1.0)
>>>         print(m.weight)
>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

property automatic_optimization: bool

If set to False you are responsible for calling .backward(), .step(), .zero_grad().

backward(loss: Tensor, optimizer: Optional[Steppable], optimizer_idx: Optional[int], *args: Any, **kwargs: Any) → None

Called to perform backward on the loss returned in training_step(). Override this hook with your own implementation if you need to.

Parameters

• loss – The loss tensor returned by training_step(). If gradient accumulation is used, the loss here holds the normalized value (scaled by 1 / accumulation steps).

• optimizer – Current optimizer being used. None if using manual optimization.

• optimizer_idx – Index of the current optimizer being used. None if using manual optimization.

Example:

def backward(self, loss, optimizer, optimizer_idx):
    loss.backward()

bfloat16() → T

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns

self

Return type

Module

buffers(recurse: bool = True) → Iterator[Tensor]

Returns an iterator over module buffers.

Parameters

recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

torch.Tensor – module buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for buf in model.buffers():
>>>     print(type(buf), buf.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

children() → Iterator[Module]

Returns an iterator over immediate children modules.

Yields

Module – a child module

clip_gradients(optimizer: Optimizer, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Handles gradient clipping internally.

Note

Do not override this method. If you want to customize gradient clipping, consider using configure_gradient_clipping() method.

Parameters

• optimizer – Current optimizer being used.

• gradient_clip_val – The value at which to clip gradients.

• gradient_clip_algorithm – The gradient clipping algorithm to use. Pass gradient_clip_algorithm="value" to clip by value, and gradient_clip_algorithm="norm" to clip by norm.

configure_gradient_clipping(optimizer: Optimizer, optimizer_idx: int, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Perform gradient clipping for the optimizer parameters. Called before optimizer_step().

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

• gradient_clip_val – The value at which to clip gradients. By default value passed in Trainer will be available here.

• gradient_clip_algorithm – The gradient clipping algorithm to use. By default value passed in Trainer will be available here.

Example:

# Perform gradient clipping on gradients associated with discriminator (optimizer_idx=1) in GAN
def configure_gradient_clipping(self, optimizer, optimizer_idx, gradient_clip_val, gradient_clip_algorithm):
    if optimizer_idx == 1:
        # Lightning will handle the gradient clipping
        self.clip_gradients(
            optimizer,
            gradient_clip_val=gradient_clip_val,
            gradient_clip_algorithm=gradient_clip_algorithm
        )
    else:
        # implement your own custom logic to clip gradients for generator (optimizer_idx=0)

configure_optimizers()

Collects learnable parameters and configures the optimizer and learning rate scheduler. :returns: two lists containing the optimizer and the scheduler. :rtype: Tuple[List, List]

configure_sharded_model() → None

Hook to create modules in a distributed aware context. This is useful for when using sharded plugins, where we’d like to shard the model instantly, which is useful for extremely large models which can save memory and initialization time.

This hook is called during each of fit/val/test/predict stages in the same process, so ensure that implementation of this hook is idempotent.

cpu() → Self

See torch.nn.Module.cpu().

cuda(device: Optional[Union[device, int]] = None) → Self

Moves all model parameters and buffers to the GPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized.

Parameters

device – If specified, all parameters will be copied to that device. If None, the current CUDA device index will be used.

Returns

self

Return type

Module

property current_epoch: int

The current epoch in the Trainer, or 0 if not attached.

double() → Self

See torch.nn.Module.double().

eval() → T

Sets the module in evaluation mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False).

See Locally disabling gradient computation for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns

self

Return type

Module

property example_input_array: Optional[Union[Tensor, Tuple, Dict]]

The example input array is a specification of what the module can consume in the forward() method. The return type is interpreted as follows:

• Single tensor: It is assumed the model takes a single argument, i.e., model.forward(model.example_input_array)

• Tuple: The input array should be interpreted as a sequence of positional arguments, i.e., model.forward(*model.example_input_array)

• Dict: The input array represents named keyword arguments, i.e., model.forward(**model.example_input_array)

extra_repr() → str

Set the extra representation of the module

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

abstract fit(views: Iterable[ndarray], y=None, **kwargs)

Fits the model to the given data

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• y (None) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

self

Return type

object

fit_transform(views: Iterable[ndarray], **kwargs)

Fits the model to the given data and returns the transformed views

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

transformed_views

Return type

list of numpy arrays

float() → Self

See torch.nn.Module.float().

freeze() → None

Freeze all params for inference.

Example:

model = MyLightningModule(...)
model.freeze()

get_buffer(target: str) → Tensor

Returns the buffer given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The buffer referenced by target

Return type

torch.Tensor

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state() → Any

Returns any extra state to include in the module’s state_dict. Implement this and a corresponding set_extra_state() for your module if you need to store extra state. This function is called when building the module’s state_dict().

Note that extra state should be pickleable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns

Any extra state to store in the module’s state_dict

Return type

object

get_factor_loadings(views: Iterable[ndarray], normalize=True, **kwargs)

Returns the factor loadings for each view

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• normalize (bool, optional) – Whether to normalize the factor loadings. Default is True.

• kwargs (any additional keyword arguments required by the given model) –

Returns

factor_loadings

Return type

list of numpy arrays

get_parameter(target: str) → Parameter

Returns the parameter given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The Parameter referenced by target

Return type

torch.nn.Parameter

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_params(deep=True)

Get parameters for this estimator.

Parameters

deep (bool, default=True) – If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns

params – Parameter names mapped to their values.

Return type

dict

get_submodule(target: str) → Module

Returns the submodule given by target if it exists, otherwise throws an error.

For example, let’s say you have an nn.Module A that looks like this:

A(
    (net_b): Module(
        (net_c): Module(
            (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2))
        )
        (linear): Linear(in_features=100, out_features=200, bias=True)
    )
)

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Parameters

target – The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns

The submodule referenced by target

Return type

torch.nn.Module

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Module

property global_rank: int

The index of the current process across all nodes and devices.

property global_step: int

Total training batches seen across all epochs.

If no Trainer is attached, this propery is 0.

half() → Self

See torch.nn.Module.half().

property hparams: Union[AttributeDict, MutableMapping]

The collection of hyperparameters saved with save_hyperparameters(). It is mutable by the user. For the frozen set of initial hyperparameters, use hparams_initial.

Returns

Mutable hyperparameters dictionary

property hparams_initial: AttributeDict

The collection of hyperparameters saved with save_hyperparameters(). These contents are read-only. Manual updates to the saved hyperparameters can instead be performed through hparams.

Returns

immutable initial hyperparameters

Return type

AttributeDict

ipu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the IPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on IPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

classmethod load_from_checkpoint(checkpoint_path: Union[str, Path, IO], map_location: Optional[Union[device, str, int, Callable[[Union[device, str, int]], Union[device, str, int]], Dict[Union[device, str, int], Union[device, str, int]]]] = None, hparams_file: Optional[Union[str, Path]] = None, strict: bool = True, **kwargs: Any) → Self

Primary way of loading a model from a checkpoint. When Lightning saves a checkpoint it stores the arguments passed to __init__ in the checkpoint under "hyper_parameters".

Any arguments specified through **kwargs will override args stored in "hyper_parameters".

Parameters

• checkpoint_path – Path to checkpoint. This can also be a URL, or file-like object

• map_location – If your checkpoint saved a GPU model and you now load on CPUs or a different number of GPUs, use this to map to the new setup. The behaviour is the same as in torch.load().

• hparams_file –

  Optional path to a .yaml or .csv file with hierarchical structure as in this example:

  drop_prob: 0.2
  dataloader:
      batch_size: 32
  

  You most likely won’t need this since Lightning will always save the hyperparameters to the checkpoint. However, if your checkpoint weights don’t have the hyperparameters saved, use this method to pass in a .yaml file with the hparams you’d like to use. These will be converted into a dict and passed into your LightningModule for use.

  If your model’s hparams argument is Namespace and .yaml file has hierarchical structure, you need to refactor your model to treat hparams as dict.

• strict – Whether to strictly enforce that the keys in checkpoint_path match the keys returned by this module’s state dict.

• **kwargs – Any extra keyword args needed to init the model. Can also be used to override saved hyperparameter values.

Returns

LightningModule instance with loaded weights and hyperparameters (if available).

Note

load_from_checkpoint is a class method. You should use your LightningModule class to call it instead of the LightningModule instance.

Example:

# load weights without mapping ...
model = MyLightningModule.load_from_checkpoint('path/to/checkpoint.ckpt')

# or load weights mapping all weights from GPU 1 to GPU 0 ...
map_location = {'cuda:1':'cuda:0'}
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    map_location=map_location
)

# or load weights and hyperparameters from separate files.
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    hparams_file='/path/to/hparams_file.yaml'
)

# override some of the params with new values
model = MyLightningModule.load_from_checkpoint(
    PATH,
    num_layers=128,
    pretrained_ckpt_path=NEW_PATH,
)

# predict
pretrained_model.eval()
pretrained_model.freeze()
y_hat = pretrained_model(x)

load_state_dict(state_dict: Mapping[str, Any], strict: bool = True)

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Parameters

• state_dict (dict) – a dict containing parameters and persistent buffers.

• strict (bool, optional) – whether to strictly enforce that the keys in state_dict match the keys returned by this module’s state_dict() function. Default: True

Returns

• missing_keys is a list of str containing the missing keys

• unexpected_keys is a list of str containing the unexpected keys

Return type

NamedTuple with missing_keys and unexpected_keys fields

Note

If a parameter or buffer is registered as None and its corresponding key exists in state_dict, load_state_dict() will raise a RuntimeError.

property local_rank: int

The index of the current process within a single node.

log(name: str, value: Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, metric_attribute: Optional[str] = None, rank_zero_only: bool = False) → None

Log a key, value pair.

Example:

self.log('train_loss', loss)

The default behavior per hook is documented here: extensions/logging:Automatic Logging.

Parameters

• name – key to log.

• value – value to log. Can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress bar.

• logger – if True logs to the logger.

• on_step – if True logs at this step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto detach the graph.

• sync_dist – if True, reduces the metric across devices. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the DDP group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple dataloaders). If False, user needs to give unique names for each dataloader to not mix the values.

• batch_size – Current batch_size. This will be directly inferred from the loaded batch, but for some data structures you might need to explicitly provide it.

• metric_attribute – To restore the metric state, Lightning requires the reference of the torchmetrics.Metric in your model. This is found automatically if it is a model attribute.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_dict(dictionary: Mapping[str, Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, rank_zero_only: bool = False) → None

Log a dictionary of values at once.

Example:

values = {'loss': loss, 'acc': acc, ..., 'metric_n': metric_n}
self.log_dict(values)

Parameters

• dictionary – key value pairs. The values can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress base.

• logger – if True logs to the logger.

• on_step – if True logs at this step. None auto-logs for training_step but not validation/test_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. None auto-logs for val/test step but not training_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto-detach the graph

• sync_dist – if True, reduces the metric across GPUs/TPUs. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the ddp group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple). If False, user needs to give unique names for each dataloader to not mix values.

• batch_size – Current batch size. This will be directly inferred from the loaded batch, but some data structures might need to explicitly provide it.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_grad_norm(grad_norm_dict: Dict[str, float]) → None

Override this method to change the default behaviour of log_grad_norm.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

grad_norm_dict – Dictionary containing current grad norm metrics

Example:

# DEFAULT
def log_grad_norm(self, grad_norm_dict):
    self.log_dict(grad_norm_dict, on_step=True, on_epoch=True, prog_bar=False, logger=True)

property logger: Optional[Logger]

Reference to the logger object in the Trainer.

property loggers: List[Logger]

Reference to the list of loggers in the Trainer.

lr_scheduler_step(scheduler: Union[_LRScheduler, ReduceLROnPlateau], optimizer_idx: int, metric: Optional[Any]) → None

Override this method to adjust the default way the Trainer calls each scheduler. By default, Lightning calls step() and as shown in the example for each scheduler based on its interval.

Parameters

• scheduler – Learning rate scheduler.

• optimizer_idx – Index of the optimizer associated with this scheduler.

• metric – Value of the monitor used for schedulers like ReduceLROnPlateau.

Examples:

# DEFAULT
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    if metric is None:
        scheduler.step()
    else:
        scheduler.step(metric)

# Alternative way to update schedulers if it requires an epoch value
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    scheduler.step(epoch=self.current_epoch)

lr_schedulers() → Union[None, List[Union[_LRScheduler, ReduceLROnPlateau]], _LRScheduler, ReduceLROnPlateau]

Returns the learning rate scheduler(s) that are being used during training. Useful for manual optimization.

Returns

A single scheduler, or a list of schedulers in case multiple ones are present, or None if no schedulers were returned in configure_optimizers().

manual_backward(loss: Tensor, *args: Any, **kwargs: Any) → None

Call this directly from your training_step() when doing optimizations manually. By using this, Lightning can ensure that all the proper scaling gets applied when using mixed precision.

See manual optimization for more examples.

Example:

def training_step(...):
    opt = self.optimizers()
    loss = ...
    opt.zero_grad()
    # automatically applies scaling, etc...
    self.manual_backward(loss)
    opt.step()

Parameters

• loss – The tensor on which to compute gradients. Must have a graph attached.

• *args – Additional positional arguments to be forwarded to backward()

• **kwargs – Additional keyword arguments to be forwarded to backward()

modules() → Iterator[Module]

Returns an iterator over all modules in the network.

Yields

Module – a module in the network

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.modules()):
...     print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Tensor]]

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Parameters

• prefix (str) – prefix to prepend to all buffer names.

• recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

(str, torch.Tensor) – Tuple containing the name and buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, buf in self.named_buffers():
>>>    if name in ['running_var']:
>>>        print(buf.size())

named_children() → Iterator[Tuple[str, Module]]

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields

(str, Module) – Tuple containing a name and child module

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, module in model.named_children():
>>>     if name in ['conv4', 'conv5']:
>>>         print(module)

named_modules(memo: Optional[Set[Module]] = None, prefix: str = '', remove_duplicate: bool = True)

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Parameters

• memo – a memo to store the set of modules already added to the result

• prefix – a prefix that will be added to the name of the module

• remove_duplicate – whether to remove the duplicated module instances in the result or not

Yields

(str, Module) – Tuple of name and module

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.named_modules()):
...     print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Parameter]]

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Parameters

• prefix (str) – prefix to prepend to all parameter names.

• recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

(str, Parameter) – Tuple containing the name and parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, param in self.named_parameters():
>>>    if name in ['bias']:
>>>        print(param.size())

on_after_backward() → None

Called after loss.backward() and before optimizers are stepped.

Note

If using native AMP, the gradients will not be unscaled at this point. Use the on_before_optimizer_step if you need the unscaled gradients.

on_after_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch after it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_after_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = gpu_transforms(batch['x'])
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• on_before_batch_transfer()

• transfer_batch_to_device()

on_before_backward(loss: Tensor) → None

Called before loss.backward().

Parameters

loss – Loss divided by number of batches for gradient accumulation and scaled if using native AMP.

on_before_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch before it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_before_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = transforms(batch['x'])
    return batch

See also

• on_after_batch_transfer()

• transfer_batch_to_device()

on_before_optimizer_step(optimizer: Optimizer, optimizer_idx: int) → None

Called before optimizer.step().

If using gradient accumulation, the hook is called once the gradients have been accumulated. See: :paramref:`~pytorch_lightning.trainer.Trainer.accumulate_grad_batches`.

If using native AMP, the loss will be unscaled before calling this hook. See these docs for more information on the scaling of gradients.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

Example:

def on_before_optimizer_step(self, optimizer, optimizer_idx):
    # example to inspect gradient information in tensorboard
    if self.trainer.global_step % 25 == 0:  # don't make the tf file huge
        for k, v in self.named_parameters():
            self.logger.experiment.add_histogram(
                tag=k, values=v.grad, global_step=self.trainer.global_step
            )

on_before_zero_grad(optimizer: Optimizer) → None

Called after training_step() and before optimizer.zero_grad().

Called in the training loop after taking an optimizer step and before zeroing grads. Good place to inspect weight information with weights updated.

This is where it is called:

for optimizer in optimizers:
    out = training_step(...)

    model.on_before_zero_grad(optimizer) # < ---- called here
    optimizer.zero_grad()

    backward()

Parameters

optimizer – The optimizer for which grads should be zeroed.

on_fit_end() → None

Called at the very end of fit.

If on DDP it is called on every process

on_fit_start() → None

Called at the very beginning of fit.

If on DDP it is called on every process

property on_gpu: bool

Returns True if this model is currently located on a GPU.

Useful to set flags around the LightningModule for different CPU vs GPU behavior.

on_load_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning to restore your model. If you saved something with on_save_checkpoint() this is your chance to restore this.

Parameters

checkpoint – Loaded checkpoint

Example:

def on_load_checkpoint(self, checkpoint):
    # 99% of the time you don't need to implement this method
    self.something_cool_i_want_to_save = checkpoint['something_cool_i_want_to_save']

Note

Lightning auto-restores global step, epoch, and train state including amp scaling. There is no need for you to restore anything regarding training.

on_predict_batch_end(outputs: Optional[Any], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop after the batch.

Parameters

• outputs – The outputs of predict_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_end() → None

Called at the end of predicting.

on_predict_epoch_end(results: List[Any]) → None

Called at the end of predicting.

on_predict_epoch_start() → None

Called at the beginning of predicting.

on_predict_model_eval() → None

Sets the model to eval during the predict loop.

on_predict_start() → None

Called at the beginning of predicting.

on_save_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning when saving a checkpoint to give you a chance to store anything else you might want to save.

Parameters

checkpoint – The full checkpoint dictionary before it gets dumped to a file. Implementations of this hook can insert additional data into this dictionary.

Example:

def on_save_checkpoint(self, checkpoint):
    # 99% of use cases you don't need to implement this method
    checkpoint['something_cool_i_want_to_save'] = my_cool_pickable_object

Note

Lightning saves all aspects of training (epoch, global step, etc…) including amp scaling. There is no need for you to store anything about training.

on_test_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop after the batch.

Parameters

• outputs – The outputs of test_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_end() → None

Called at the end of testing.

on_test_epoch_end() → None

Called in the test loop at the very end of the epoch.

on_test_epoch_start() → None

Called in the test loop at the very beginning of the epoch.

on_test_model_eval() → None

Sets the model to eval during the test loop.

on_test_model_train() → None

Sets the model to train during the test loop.

on_test_start() → None

Called at the beginning of testing.

on_train_batch_end(outputs: Union[Tensor, Dict[str, Any]], batch: Any, batch_idx: int) → None

Called in the training loop after the batch.

Parameters

• outputs – The outputs of training_step_end(training_step(x))

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_batch_start(batch: Any, batch_idx: int) → Optional[int]

Called in the training loop before anything happens for that batch.

If you return -1 here, you will skip training for the rest of the current epoch.

Parameters

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_end() → None

Called at the end of training before logger experiment is closed.

on_train_epoch_end() → None

Called in the training loop at the very end of the epoch.

To access all batch outputs at the end of the epoch, either:

1. Implement training_epoch_end in the LightningModule OR

2. Cache data across steps on the attribute(s) of the LightningModule and access them in this hook

on_train_epoch_start() → None

Called in the training loop at the very beginning of the epoch.

on_train_start() → None

Called at the beginning of training after sanity check.

on_validation_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop after the batch.

Parameters

• outputs – The outputs of validation_step_end(validation_step(x))

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_end() → None

Called at the end of validation.

on_validation_epoch_end() → None

Called in the validation loop at the very end of the epoch.

on_validation_epoch_start() → None

Called in the validation loop at the very beginning of the epoch.

on_validation_model_eval() → None

Sets the model to eval during the val loop.

on_validation_model_train() → None

Sets the model to train during the val loop.

on_validation_start() → None

Called at the beginning of validation.

optimizer_step(epoch: int, batch_idx: int, optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int = 0, optimizer_closure: Optional[Callable[[], Any]] = None, on_tpu: bool = False, using_native_amp: bool = False, using_lbfgs: bool = False) → None

Override this method to adjust the default way the Trainer calls each optimizer.

By default, Lightning calls step() and zero_grad() as shown in the example once per optimizer. This method (and zero_grad()) won’t be called during the accumulation phase when Trainer(accumulate_grad_batches != 1). Overriding this hook has no benefit with manual optimization.

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers, this indexes into that list.

• optimizer_closure – The optimizer closure. This closure must be executed as it includes the calls to training_step(), optimizer.zero_grad(), and backward().

• on_tpu – True if TPU backward is required

• using_native_amp – True if using native amp

• using_lbfgs – True if the matching optimizer is torch.optim.LBFGS

Examples:

# DEFAULT
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    optimizer.step(closure=optimizer_closure)

# Alternating schedule for optimizer steps (i.e.: GANs)
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    # update generator opt every step
    if optimizer_idx == 0:
        optimizer.step(closure=optimizer_closure)

    # update discriminator opt every 2 steps
    if optimizer_idx == 1:
        if (batch_idx + 1) % 2 == 0 :
            optimizer.step(closure=optimizer_closure)
        else:
            # call the closure by itself to run `training_step` + `backward` without an optimizer step
            optimizer_closure()

    # ...
    # add as many optimizers as you want

Here’s another example showing how to use this for more advanced things such as learning rate warm-up:

# learning rate warm-up
def optimizer_step(
    self,
    epoch,
    batch_idx,
    optimizer,
    optimizer_idx,
    optimizer_closure,
    on_tpu,
    using_native_amp,
    using_lbfgs,
):
    # update params
    optimizer.step(closure=optimizer_closure)

    # manually warm up lr without a scheduler
    if self.trainer.global_step < 500:
        lr_scale = min(1.0, float(self.trainer.global_step + 1) / 500.0)
        for pg in optimizer.param_groups:
            pg["lr"] = lr_scale * self.learning_rate

optimizer_zero_grad(epoch: int, batch_idx: int, optimizer: Optimizer, optimizer_idx: int) → None

Override this method to change the default behaviour of optimizer.zero_grad().

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers this indexes into that list.

Examples:

# DEFAULT
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad()

# Set gradients to `None` instead of zero to improve performance.
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad(set_to_none=True)

See torch.optim.Optimizer.zero_grad() for the explanation of the above example.

optimizers(use_pl_optimizer: bool = True) → Union[Optimizer, LightningOptimizer, List[Optimizer], List[LightningOptimizer]]

Returns the optimizer(s) that are being used during training. Useful for manual optimization.

Parameters

use_pl_optimizer – If True, will wrap the optimizer(s) in a LightningOptimizer for automatic handling of precision and profiling.

Returns

A single optimizer, or a list of optimizers in case multiple ones are present.

pairwise_correlations(loader: DataLoader, train=False)

Calculates correlation for entire batch from dataloader

Parameters

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

Returns

by default returns the average pairwise correlation in each dimension (for 2 views just the correlation)

parameters(recurse: bool = True) → Iterator[Parameter]

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Parameters

recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

Parameter – module parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for param in model.parameters():
>>>     print(type(param), param.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

predict_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for prediction.

It’s recommended that all data downloads and preparation happen in prepare_data().

• predict()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying prediction samples.

Note

In the case where you return multiple prediction dataloaders, the predict_step() will have an argument dataloader_idx which matches the order here.

predict_step(batch: Any, batch_idx: int, dataloader_idx: int = 0) → Any

Step function called during predict(). By default, it calls forward(). Override to add any processing logic.

The predict_step() is used to scale inference on multi-devices.

To prevent an OOM error, it is possible to use BasePredictionWriter callback to write the predictions to disk or database after each batch or on epoch end.

The BasePredictionWriter should be used while using a spawn based accelerator. This happens for Trainer(strategy="ddp_spawn") or training on 8 TPU cores with Trainer(accelerator="tpu", devices=8) as predictions won’t be returned.

Example

class MyModel(LightningModule):

    def predict_step(self, batch, batch_idx, dataloader_idx=0):
        return self(batch)

dm = ...
model = MyModel()
trainer = Trainer(accelerator="gpu", devices=2)
predictions = trainer.predict(model, dm)

Parameters

• batch – Current batch.

• batch_idx – Index of current batch.

• dataloader_idx – Index of the current dataloader.

Returns

Predicted output

prepare_data() → None

Use this to download and prepare data. Downloading and saving data with multiple processes (distributed settings) will result in corrupted data. Lightning ensures this method is called only within a single process, so you can safely add your downloading logic within.

Warning

DO NOT set state to the model (use setup instead) since this is NOT called on every device

Example:

def prepare_data(self):
    # good
    download_data()
    tokenize()
    etc()

    # bad
    self.split = data_split
    self.some_state = some_other_state()

In a distributed environment, prepare_data can be called in two ways (using prepare_data_per_node)

1. Once per node. This is the default and is only called on LOCAL_RANK=0.

2. Once in total. Only called on GLOBAL_RANK=0.

Example:

# DEFAULT
# called once per node on LOCAL_RANK=0 of that node
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = True


# call on GLOBAL_RANK=0 (great for shared file systems)
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = False

This is called before requesting the dataloaders:

model.prepare_data()
initialize_distributed()
model.setup(stage)
model.train_dataloader()
model.val_dataloader()
model.test_dataloader()
model.predict_dataloader()

print(*args: Any, **kwargs: Any) → None

Prints only from process 0. Use this in any distributed mode to log only once.

Parameters

• *args – The thing to print. The same as for Python’s built-in print function.

• **kwargs – The same as for Python’s built-in print function.

Example:

def forward(self, x):
    self.print(x, 'in forward')

register_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

This function is deprecated in favor of register_full_backward_hook() and the behavior of this function will change in future versions.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_buffer(name: str, tensor: Optional[Tensor], persistent: bool = True) → None

Adds a buffer to the module.

This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the module’s state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module’s state_dict.

Buffers can be accessed as attributes using given names.

Parameters

• name (str) – name of the buffer. The buffer can be accessed from this module using the given name

• tensor (Tensor or None) – buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module’s state_dict.

• persistent (bool) – whether the buffer is part of this module’s state_dict.

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> self.register_buffer('running_mean', torch.zeros(num_features))

register_forward_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output. It should have the following signature:

hook(module, input, output) -> None or modified output

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward() is called.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_forward_pre_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked. It should have the following signature:

hook(module, input) -> None or modified input

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned(unless that value is already a tuple).

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_full_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

The hook will be called every time the gradients with respect to module inputs are computed. The hook should have the following signature:

hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module’s forward function.

Warning

Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_load_state_dict_post_hook(hook)

Registers a post hook to be run after module’s load_state_dict is called.

It should have the following signature::

hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict() with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearning out both missing and unexpected keys will avoid an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_module(name: str, module: Optional[Module]) → None

Alias for add_module().

register_parameter(name: str, param: Optional[Parameter]) → None

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Parameters

• name (str) – name of the parameter. The parameter can be accessed from this module using the given name

• param (Parameter or None) – parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module’s state_dict.

requires_grad_(requires_grad: bool = True) → T

Change if autograd should record operations on parameters in this module.

This method sets the parameters’ requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See Locally disabling gradient computation for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Parameters

requires_grad (bool) – whether autograd should record operations on parameters in this module. Default: True.

Returns

self

Return type

Module

save_hyperparameters(*args: Any, ignore: Optional[Union[Sequence[str], str]] = None, frame: Optional[frame] = None, logger: bool = True) → None

Save arguments to hparams attribute.

Parameters

• args – single object of dict, NameSpace or OmegaConf or string names or arguments from class __init__

• ignore – an argument name or a list of argument names from class __init__ to be ignored

• frame – a frame object. Default is None

• logger – Whether to send the hyperparameters to the logger. Default: True

Example::

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # manually assign arguments
...         self.save_hyperparameters('arg1', 'arg3')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

>>> class AutomaticArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # equivalent automatic
...         self.save_hyperparameters()
...     def forward(self, *args, **kwargs):
...         ...
>>> model = AutomaticArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg2": abc
"arg3": 3.14

>>> class SingleArgModel(HyperparametersMixin):
...     def __init__(self, params):
...         super().__init__()
...         # manually assign single argument
...         self.save_hyperparameters(params)
...     def forward(self, *args, **kwargs):
...         ...
>>> model = SingleArgModel(Namespace(p1=1, p2='abc', p3=3.14))
>>> model.hparams
"p1": 1
"p2": abc
"p3": 3.14

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # pass argument(s) to ignore as a string or in a list
...         self.save_hyperparameters(ignore='arg2')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

score(loader: DataLoader, **kwargs)

Returns average correlation in each dimension (averages over all pairs for multiview)

Parameters

• **kwargs –

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

set_extra_state(state: Any)

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Parameters

state (dict) – Extra state from the state_dict

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters

**params (dict) – Estimator parameters.

Returns

self – Estimator instance.

Return type

estimator instance

setup(stage: str) → None

Called at the beginning of fit (train + validate), validate, test, or predict. This is a good hook when you need to build models dynamically or adjust something about them. This hook is called on every process when using DDP.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

Example:

class LitModel(...):
    def __init__(self):
        self.l1 = None

    def prepare_data(self):
        download_data()
        tokenize()

        # don't do this
        self.something = else

    def setup(self, stage):
        data = load_data(...)
        self.l1 = nn.Linear(28, data.num_classes)

share_memory() → T

See torch.Tensor.share_memory_()

state_dict(*args, destination=None, prefix='', keep_vars=False)

Returns a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

Note

The returned object is a shallow copy. It contains references to the module’s parameters and buffers.

Warning

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

Warning

Please avoid the use of argument destination as it is not designed for end-users.

Parameters

• destination (dict, optional) – If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned. Default: None.

• prefix (str, optional) – a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.

• keep_vars (bool, optional) – by default the Tensor s returned in the state dict are detached from autograd. If it’s set to True, detaching will not be performed. Default: False.

Returns

a dictionary containing a whole state of the module

Return type

dict

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> module.state_dict().keys()
['bias', 'weight']

tbptt_split_batch(batch: Any, split_size: int) → List[Any]

When using truncated backpropagation through time, each batch must be split along the time dimension. Lightning handles this by default, but for custom behavior override this function.

Parameters

• batch – Current batch

• split_size – The size of the split

Returns

List of batch splits. Each split will be passed to training_step() to enable truncated back propagation through time. The default implementation splits root level Tensors and Sequences at dim=1 (i.e. time dim). It assumes that each time dim is the same length.

Examples:

def tbptt_split_batch(self, batch, split_size):
    splits = []
    for t in range(0, time_dims[0], split_size):
        batch_split = []
        for i, x in enumerate(batch):
            if isinstance(x, torch.Tensor):
                split_x = x[:, t:t + split_size]
            elif isinstance(x, collections.abc.Sequence):
                split_x = [None] * len(x)
                for batch_idx in range(len(x)):
                  split_x[batch_idx] = x[batch_idx][t:t + split_size]
            batch_split.append(split_x)
        splits.append(batch_split)
    return splits

Note

Called in the training loop after on_train_batch_start() if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0. Each returned batch split is passed separately to training_step().

teardown(stage: str) → None

Called at the end of fit (train + validate), validate, test, or predict.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

test_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for testing.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• test()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying testing samples.

Example:

def test_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def test_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a test dataset and a test_step(), you don’t need to implement this method.

Note

In the case where you return multiple test dataloaders, the test_step() will have an argument dataloader_idx which matches the order here.

test_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of a test epoch with the output of all test steps.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

outputs – List of outputs you defined in test_step_end(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader

Returns

None

Note

If you didn’t define a test_step(), this won’t be called.

Examples

With a single dataloader:

def test_epoch_end(self, outputs):
    # do something with the outputs of all test batches
    all_test_preds = test_step_outputs.predictions

    some_result = calc_all_results(all_test_preds)
    self.log(some_result)

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each test step for that dataloader.

def test_epoch_end(self, outputs):
    final_value = 0
    for dataloader_outputs in outputs:
        for test_step_out in dataloader_outputs:
            # do something
            final_value += test_step_out

    self.log("final_metric", final_value)

test_step(batch, batch_idx)

Operates on a single batch of data from the test set. In this step you’d normally generate examples or calculate anything of interest such as accuracy.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_id – The index of the dataloader that produced this batch. (only if multiple test dataloaders used).

Returns

Any of.

• Any object or value

• None - Testing will skip to the next batch

# if you have one test dataloader:
def test_step(self, batch, batch_idx):
    ...


# if you have multiple test dataloaders:
def test_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single test dataset
def test_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    test_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'test_loss': loss, 'test_acc': test_acc})

If you pass in multiple test dataloaders, test_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple test dataloaders
def test_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to test you don’t need to implement this method.

Note

When the test_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of the test epoch, the model goes back to training mode and gradients are enabled.

test_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when testing with DP because test_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [test_step(sub_batch) for sub_batch in sub_batches]
test_step_end(step_output)

Parameters

step_output – What you return in test_step() for each batch part.

Returns

None or anything

# WITHOUT test_step_end
# if used in DP, this batch is 1/num_gpus large
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    loss = self.softmax(out)
    self.log("test_loss", loss)


# --------------
# with test_step_end to do softmax over the full batch
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return out


def test_step_end(self, output_results):
    # this out is now the full size of the batch
    all_test_step_outs = output_results.out
    loss = nce_loss(all_test_step_outs)
    self.log("test_loss", loss)

See also

See the Multi GPU Training guide for more details.

to(*args: Any, **kwargs: Any) → Self

See torch.nn.Module.to().

to_empty(*, device: Union[str, device]) → T

Moves the parameters and buffers to the specified device without copying storage.

Parameters

device (torch.device) – The desired device of the parameters and buffers in this module.

Returns

self

Return type

Module

to_onnx(file_path: Union[str, Path], input_sample: Optional[Any] = None, **kwargs: Any) → None

Saves the model in ONNX format.

Parameters

• file_path – The path of the file the onnx model should be saved to.

• input_sample – An input for tracing. Default: None (Use self.example_input_array)

• **kwargs – Will be passed to torch.onnx.export function.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))

>>> with tempfile.NamedTemporaryFile(suffix='.onnx', delete=False) as tmpfile:
...     model = SimpleModel()
...     input_sample = torch.randn((1, 64))
...     model.to_onnx(tmpfile.name, input_sample, export_params=True)
...     os.path.isfile(tmpfile.name)
True

to_torchscript(file_path: Optional[Union[str, Path]] = None, method: Optional[str] = 'script', example_inputs: Optional[Any] = None, **kwargs: Any) → Union[ScriptModule, Dict[str, ScriptModule]]

By default compiles the whole model to a ScriptModule. If you want to use tracing, please provided the argument method='trace' and make sure that either the example_inputs argument is provided, or the model has example_input_array set. If you would like to customize the modules that are scripted you should override this method. In case you want to return multiple modules, we recommend using a dictionary.

Parameters

• file_path – Path where to save the torchscript. Default: None (no file saved).

• method – Whether to use TorchScript’s script or trace method. Default: ‘script’

• example_inputs – An input to be used to do tracing when method is set to ‘trace’. Default: None (uses example_input_array)

• **kwargs – Additional arguments that will be passed to the torch.jit.script() or torch.jit.trace() function.

Note

• Requires the implementation of the forward() method.

• The exported script will be set to evaluation mode.

• It is recommended that you install the latest supported version of PyTorch to use this feature without limitations. See also the torch.jit documentation for supported features.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))
...
>>> model = SimpleModel()
>>> model.to_torchscript(file_path="model.pt")  
>>> os.path.isfile("model.pt")  
>>> torch.jit.save(model.to_torchscript(file_path="model_trace.pt", method='trace', 
...                                     example_inputs=torch.randn(1, 64)))  
>>> os.path.isfile("model_trace.pt")  
True

Returns

This LightningModule as a torchscript, regardless of whether file_path is defined or not.

toggle_optimizer(optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int) → None

Makes sure only the gradients of the current optimizer’s parameters are calculated in the training step to prevent dangling gradients in multiple-optimizer setup.

This is only called automatically when automatic optimization is enabled and multiple optimizers are used. It works with untoggle_optimizer() to make sure param_requires_grad_state is properly reset.

Parameters

• optimizer – The optimizer to toggle.

• optimizer_idx – The index of the optimizer to toggle.

train(mode: bool = True) → T

Sets the module in training mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

Parameters

mode (bool) – whether to set training mode (True) or evaluation mode (False). Default: True.

Returns

self

Return type

Module

train_dataloader() → Union[DataLoader, Sequence[DataLoader], Sequence[Sequence[DataLoader]], Sequence[Dict[str, DataLoader]], Dict[str, DataLoader], Dict[str, Dict[str, DataLoader]], Dict[str, Sequence[DataLoader]]]

Implement one or more PyTorch DataLoaders for training.

Returns

A collection of torch.utils.data.DataLoader specifying training samples. In the case of multiple dataloaders, please see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• fit()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Example:

# single dataloader
def train_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=True, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=True
    )
    return loader

# multiple dataloaders, return as list
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a list of tensors: [batch_mnist, batch_cifar]
    return [mnist_loader, cifar_loader]

# multiple dataloader, return as dict
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a dict of tensors: {'mnist': batch_mnist, 'cifar': batch_cifar}
    return {'mnist': mnist_loader, 'cifar': cifar_loader}

training_epoch_end(outputs: List[Union[Tensor, Dict[str, Any]]]) → None

Called at the end of the training epoch with the outputs of all training steps. Use this in case you need to do something with all the outputs returned by training_step().

# the pseudocode for these calls
train_outs = []
for train_batch in train_data:
    out = training_step(train_batch)
    train_outs.append(out)
training_epoch_end(train_outs)

Parameters

outputs – List of outputs you defined in training_step(). If there are multiple optimizers or when using truncated_bptt_steps > 0, the lists have the dimensions (n_batches, tbptt_steps, n_optimizers). Dimensions of length 1 are squeezed.

Returns

None

Note

If this method is not overridden, this won’t be called.

def training_epoch_end(self, training_step_outputs):
    # do something with all training_step outputs
    for out in training_step_outputs:
        ...

training_step(batch, batch_idx)

Here you compute and return the training loss and some additional metrics for e.g. the progress bar or logger.

Parameters

• batch (Tensor | (Tensor, …) | [Tensor, …]) – The output of your DataLoader. A tensor, tuple or list.

• batch_idx (int) – Integer displaying index of this batch

• optimizer_idx (int) – When using multiple optimizers, this argument will also be present.

• hiddens (Any) – Passed in if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0.

Returns

Any of.

• Tensor - The loss tensor

• dict - A dictionary. Can include any keys, but must include the key 'loss'

• None - Training will skip to the next batch. This is only for automatic optimization.

  This is not supported for multi-GPU, TPU, IPU, or DeepSpeed.

In this step you’d normally do the forward pass and calculate the loss for a batch. You can also do fancier things like multiple forward passes or something model specific.

Example:

def training_step(self, batch, batch_idx):
    x, y, z = batch
    out = self.encoder(x)
    loss = self.loss(out, x)
    return loss

If you define multiple optimizers, this step will be called with an additional optimizer_idx parameter.

# Multiple optimizers (e.g.: GANs)
def training_step(self, batch, batch_idx, optimizer_idx):
    if optimizer_idx == 0:
        # do training_step with encoder
        ...
    if optimizer_idx == 1:
        # do training_step with decoder
        ...

If you add truncated back propagation through time you will also get an additional argument with the hidden states of the previous step.

# Truncated back-propagation through time
def training_step(self, batch, batch_idx, hiddens):
    # hiddens are the hidden states from the previous truncated backprop step
    out, hiddens = self.lstm(data, hiddens)
    loss = ...
    return {"loss": loss, "hiddens": hiddens}

Note

The loss value shown in the progress bar is smoothed (averaged) over the last values, so it differs from the actual loss returned in train/validation step.

Note

When accumulate_grad_batches > 1, the loss returned here will be automatically normalized by accumulate_grad_batches internally.

training_step_end(step_output: Union[Tensor, Dict[str, Any]]) → Union[Tensor, Dict[str, Any]]

Use this when training with dp because training_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [training_step(sub_batch) for sub_batch in sub_batches]
training_step_end(step_output)

Parameters

step_output – What you return in training_step for each batch part.

Returns

Anything

When using the DP strategy, only a portion of the batch is inside the training_step:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)

    # softmax uses only a portion of the batch in the denominator
    loss = self.softmax(out)
    loss = nce_loss(loss)
    return loss

If you wish to do something with all the parts of the batch, then use this method to do it:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return {"pred": out}


def training_step_end(self, training_step_outputs):
    gpu_0_pred = training_step_outputs[0]["pred"]
    gpu_1_pred = training_step_outputs[1]["pred"]
    gpu_n_pred = training_step_outputs[n]["pred"]

    # this softmax now uses the full batch
    loss = nce_loss([gpu_0_pred, gpu_1_pred, gpu_n_pred])
    return loss

See also

See the Multi GPU Training guide for more details.

transfer_batch_to_device(batch: Any, device: device, dataloader_idx: int) → Any

Override this hook if your DataLoader returns tensors wrapped in a custom data structure.

The data types listed below (and any arbitrary nesting of them) are supported out of the box:

• torch.Tensor or anything that implements .to(…)

• list

• dict

• tuple

For anything else, you need to define how the data is moved to the target device (CPU, GPU, TPU, …).

Note

This hook should only transfer the data and not modify it, nor should it move the data to any other device than the one passed in as argument (unless you know what you are doing). To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be transferred to a new device.

• device – The target device as defined in PyTorch.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A reference to the data on the new device.

Example:

def transfer_batch_to_device(self, batch, device, dataloader_idx):
    if isinstance(batch, CustomBatch):
        # move all tensors in your custom data structure to the device
        batch.samples = batch.samples.to(device)
        batch.targets = batch.targets.to(device)
    elif dataloader_idx == 0:
        # skip device transfer for the first dataloader or anything you wish
        pass
    else:
        batch = super().transfer_batch_to_device(data, device, dataloader_idx)
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• move_data_to_device()

• apply_to_collection()

transform(loader: DataLoader)

Parameters

loader – a dataloader that matches the structure of that used for training

Returns

transformed views

property truncated_bptt_steps: int

Enables Truncated Backpropagation Through Time in the Trainer when set to a positive integer.

It represents the number of times training_step() gets called before backpropagation. If this is > 0, the training_step() receives an additional argument hiddens and is expected to return a hidden state.

type(dst_type: Union[str, dtype]) → Self

See torch.nn.Module.type().

unfreeze() → None

Unfreeze all parameters for training.

model = MyLightningModule(...)
model.unfreeze()

untoggle_optimizer(optimizer_idx: int) → None

Resets the state of required gradients that were toggled with toggle_optimizer().

This is only called automatically when automatic optimization is enabled and multiple optimizers are used.

Parameters

optimizer_idx – The index of the optimizer to untoggle.

val_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for validation.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

It’s recommended that all data downloads and preparation happen in prepare_data().

• fit()

• validate()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying validation samples.

Examples:

def val_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False,
                    transform=transform, download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def val_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a validation dataset and a validation_step(), you don’t need to implement this method.

Note

In the case where you return multiple validation dataloaders, the validation_step() will have an argument dataloader_idx which matches the order here.

validation_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of the validation epoch with the outputs of all validation steps.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

outputs – List of outputs you defined in validation_step(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader.

Returns

None

Note

If you didn’t define a validation_step(), this won’t be called.

Examples

With a single dataloader:

def validation_epoch_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each validation step for that dataloader.

def validation_epoch_end(self, outputs):
    for dataloader_output_result in outputs:
        dataloader_outs = dataloader_output_result.dataloader_i_outputs

    self.log("final_metric", final_value)

validation_step(batch, batch_idx)

Operates on a single batch of data from the validation set. In this step you’d might generate examples or calculate anything of interest like accuracy.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_idx – The index of the dataloader that produced this batch. (only if multiple val dataloaders used)

Returns

• Any object or value

• None - Validation will skip to the next batch

# pseudocode of order
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    if defined("validation_step_end"):
        out = validation_step_end(out)
    val_outs.append(out)
val_outs = validation_epoch_end(val_outs)

# if you have one val dataloader:
def validation_step(self, batch, batch_idx):
    ...


# if you have multiple val dataloaders:
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single validation dataset
def validation_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    val_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'val_loss': loss, 'val_acc': val_acc})

If you pass in multiple val dataloaders, validation_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple validation dataloaders
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to validate you don’t need to implement this method.

Note

When the validation_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of validation, the model goes back to training mode and gradients are enabled.

validation_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when validating with dp because validation_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [validation_step(sub_batch) for sub_batch in sub_batches]
validation_step_end(step_output)

Parameters

step_output – What you return in validation_step() for each batch part.

Returns

None or anything

# WITHOUT validation_step_end
# if used in DP, this batch is 1/num_gpus large
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    loss = self.softmax(out)
    loss = nce_loss(loss)
    self.log("val_loss", loss)


# --------------
# with validation_step_end to do softmax over the full batch
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    return out


def validation_step_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

See also

See the Multi GPU Training guide for more details.

xpu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the XPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

zero_grad(set_to_none: bool = False) → None

Sets gradients of all model parameters to zero. See similar function under torch.optim.Optimizer for more context.

Parameters

set_to_none (bool) – instead of setting to zero, set the grads to None. See torch.optim.Optimizer.zero_grad() for details.

class cca_zoo.deepmodels.DCCA_NOI(latent_dims: int, N: int, encoders=None, r: float = 0, rho: float = 0.2, eps: float = 1e-09, shared_target: bool = False, **kwargs)

Bases: DCCA

A class used to fit a DCCA model by non-linear orthogonal iterations

References

Wang, Weiran, et al. “Stochastic optimization for deep CCA via nonlinear orthogonal iterations.” 2015 53rd Annual Allerton Conference on Communication, Control, and Computing (Allerton). IEEE, 2015.

loss(views, **kwargs)

Loss for model :param views: :param args: :param kwargs: :return:

add_module(name: str, module: Optional[Module]) → None

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Parameters

• name (str) – name of the child module. The child module can be accessed from this module using the given name

• module (Module) – child module to be added to the module.

all_gather(data: Union[Tensor, Dict, List, Tuple], group: Optional[Any] = None, sync_grads: bool = False) → Union[Tensor, Dict, List, Tuple]

Allows users to call self.all_gather() from the LightningModule, thus making the all_gather operation accelerator agnostic. all_gather is a function provided by accelerators to gather a tensor from several distributed processes.

Parameters

• data – int, float, tensor of shape (batch, …), or a (possibly nested) collection thereof.

• group – the process group to gather results from. Defaults to all processes (world)

• sync_grads – flag that allows users to synchronize gradients for the all_gather operation

Returns

A tensor of shape (world_size, batch, …), or if the input was a collection the output will also be a collection with tensors of this shape.

apply(fn: Callable[[Module], None]) → T

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also torch.nn.init).

Parameters

fn (Module -> None) – function to be applied to each submodule

Returns

self

Return type

Module

Example:

>>> @torch.no_grad()
>>> def init_weights(m):
>>>     print(m)
>>>     if type(m) == nn.Linear:
>>>         m.weight.fill_(1.0)
>>>         print(m.weight)
>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

property automatic_optimization: bool

If set to False you are responsible for calling .backward(), .step(), .zero_grad().

backward(loss: Tensor, optimizer: Optional[Steppable], optimizer_idx: Optional[int], *args: Any, **kwargs: Any) → None

Called to perform backward on the loss returned in training_step(). Override this hook with your own implementation if you need to.

Parameters

• loss – The loss tensor returned by training_step(). If gradient accumulation is used, the loss here holds the normalized value (scaled by 1 / accumulation steps).

• optimizer – Current optimizer being used. None if using manual optimization.

• optimizer_idx – Index of the current optimizer being used. None if using manual optimization.

Example:

def backward(self, loss, optimizer, optimizer_idx):
    loss.backward()

bfloat16() → T

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns

self

Return type

Module

buffers(recurse: bool = True) → Iterator[Tensor]

Returns an iterator over module buffers.

Parameters

recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

torch.Tensor – module buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for buf in model.buffers():
>>>     print(type(buf), buf.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

children() → Iterator[Module]

Returns an iterator over immediate children modules.

Yields

Module – a child module

clip_gradients(optimizer: Optimizer, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Handles gradient clipping internally.

Note

Do not override this method. If you want to customize gradient clipping, consider using configure_gradient_clipping() method.

Parameters

• optimizer – Current optimizer being used.

• gradient_clip_val – The value at which to clip gradients.

• gradient_clip_algorithm – The gradient clipping algorithm to use. Pass gradient_clip_algorithm="value" to clip by value, and gradient_clip_algorithm="norm" to clip by norm.

configure_callbacks()

Configure model-specific callbacks. When the model gets attached, e.g., when .fit() or .test() gets called, the list or a callback returned here will be merged with the list of callbacks passed to the Trainer’s callbacks argument. If a callback returned here has the same type as one or several callbacks already present in the Trainer’s callbacks list, it will take priority and replace them. In addition, Lightning will make sure ModelCheckpoint callbacks run last.

Returns

A callback or a list of callbacks which will extend the list of callbacks in the Trainer.

Example:

def configure_callbacks(self):
    early_stop = EarlyStopping(monitor="val_acc", mode="max")
    checkpoint = ModelCheckpoint(monitor="val_loss")
    return [early_stop, checkpoint]

configure_gradient_clipping(optimizer: Optimizer, optimizer_idx: int, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Perform gradient clipping for the optimizer parameters. Called before optimizer_step().

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

• gradient_clip_val – The value at which to clip gradients. By default value passed in Trainer will be available here.

• gradient_clip_algorithm – The gradient clipping algorithm to use. By default value passed in Trainer will be available here.

Example:

# Perform gradient clipping on gradients associated with discriminator (optimizer_idx=1) in GAN
def configure_gradient_clipping(self, optimizer, optimizer_idx, gradient_clip_val, gradient_clip_algorithm):
    if optimizer_idx == 1:
        # Lightning will handle the gradient clipping
        self.clip_gradients(
            optimizer,
            gradient_clip_val=gradient_clip_val,
            gradient_clip_algorithm=gradient_clip_algorithm
        )
    else:
        # implement your own custom logic to clip gradients for generator (optimizer_idx=0)

configure_optimizers()

Collects learnable parameters and configures the optimizer and learning rate scheduler. :returns: two lists containing the optimizer and the scheduler. :rtype: Tuple[List, List]

configure_sharded_model() → None

Hook to create modules in a distributed aware context. This is useful for when using sharded plugins, where we’d like to shard the model instantly, which is useful for extremely large models which can save memory and initialization time.

This hook is called during each of fit/val/test/predict stages in the same process, so ensure that implementation of this hook is idempotent.

cpu() → Self

See torch.nn.Module.cpu().

cuda(device: Optional[Union[device, int]] = None) → Self

Moves all model parameters and buffers to the GPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized.

Parameters

device – If specified, all parameters will be copied to that device. If None, the current CUDA device index will be used.

Returns

self

Return type

Module

property current_epoch: int

The current epoch in the Trainer, or 0 if not attached.

double() → Self

See torch.nn.Module.double().

eval() → T

Sets the module in evaluation mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False).

See Locally disabling gradient computation for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns

self

Return type

Module

property example_input_array: Optional[Union[Tensor, Tuple, Dict]]

The example input array is a specification of what the module can consume in the forward() method. The return type is interpreted as follows:

• Single tensor: It is assumed the model takes a single argument, i.e., model.forward(model.example_input_array)

• Tuple: The input array should be interpreted as a sequence of positional arguments, i.e., model.forward(*model.example_input_array)

• Dict: The input array represents named keyword arguments, i.e., model.forward(**model.example_input_array)

extra_repr() → str

Set the extra representation of the module

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

abstract fit(views: Iterable[ndarray], y=None, **kwargs)

Fits the model to the given data

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• y (None) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

self

Return type

object

fit_transform(views: Iterable[ndarray], **kwargs)

Fits the model to the given data and returns the transformed views

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

transformed_views

Return type

list of numpy arrays

float() → Self

See torch.nn.Module.float().

forward(views, **kwargs)

We use the forward model to define the transformation of views to the latent space :param views: batches for each view separated by commas

freeze() → None

Freeze all params for inference.

Example:

model = MyLightningModule(...)
model.freeze()

get_buffer(target: str) → Tensor

Returns the buffer given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The buffer referenced by target

Return type

torch.Tensor

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state() → Any

Returns any extra state to include in the module’s state_dict. Implement this and a corresponding set_extra_state() for your module if you need to store extra state. This function is called when building the module’s state_dict().

Note that extra state should be pickleable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns

Any extra state to store in the module’s state_dict

Return type

object

get_factor_loadings(views: Iterable[ndarray], normalize=True, **kwargs)

Returns the factor loadings for each view

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• normalize (bool, optional) – Whether to normalize the factor loadings. Default is True.

• kwargs (any additional keyword arguments required by the given model) –

Returns

factor_loadings

Return type

list of numpy arrays

get_parameter(target: str) → Parameter

Returns the parameter given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The Parameter referenced by target

Return type

torch.nn.Parameter

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_params(deep=True)

Get parameters for this estimator.

Parameters

deep (bool, default=True) – If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns

params – Parameter names mapped to their values.

Return type

dict

get_submodule(target: str) → Module

Returns the submodule given by target if it exists, otherwise throws an error.

For example, let’s say you have an nn.Module A that looks like this:

A(
    (net_b): Module(
        (net_c): Module(
            (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2))
        )
        (linear): Linear(in_features=100, out_features=200, bias=True)
    )
)

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Parameters

target – The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns

The submodule referenced by target

Return type

torch.nn.Module

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Module

property global_rank: int

The index of the current process across all nodes and devices.

property global_step: int

Total training batches seen across all epochs.

If no Trainer is attached, this propery is 0.

half() → Self

See torch.nn.Module.half().

property hparams: Union[AttributeDict, MutableMapping]

The collection of hyperparameters saved with save_hyperparameters(). It is mutable by the user. For the frozen set of initial hyperparameters, use hparams_initial.

Returns

Mutable hyperparameters dictionary

property hparams_initial: AttributeDict

The collection of hyperparameters saved with save_hyperparameters(). These contents are read-only. Manual updates to the saved hyperparameters can instead be performed through hparams.

Returns

immutable initial hyperparameters

Return type

AttributeDict

ipu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the IPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on IPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

classmethod load_from_checkpoint(checkpoint_path: Union[str, Path, IO], map_location: Optional[Union[device, str, int, Callable[[Union[device, str, int]], Union[device, str, int]], Dict[Union[device, str, int], Union[device, str, int]]]] = None, hparams_file: Optional[Union[str, Path]] = None, strict: bool = True, **kwargs: Any) → Self

Primary way of loading a model from a checkpoint. When Lightning saves a checkpoint it stores the arguments passed to __init__ in the checkpoint under "hyper_parameters".

Any arguments specified through **kwargs will override args stored in "hyper_parameters".

Parameters

• checkpoint_path – Path to checkpoint. This can also be a URL, or file-like object

• map_location – If your checkpoint saved a GPU model and you now load on CPUs or a different number of GPUs, use this to map to the new setup. The behaviour is the same as in torch.load().

• hparams_file –

  Optional path to a .yaml or .csv file with hierarchical structure as in this example:

  drop_prob: 0.2
  dataloader:
      batch_size: 32
  

  You most likely won’t need this since Lightning will always save the hyperparameters to the checkpoint. However, if your checkpoint weights don’t have the hyperparameters saved, use this method to pass in a .yaml file with the hparams you’d like to use. These will be converted into a dict and passed into your LightningModule for use.

  If your model’s hparams argument is Namespace and .yaml file has hierarchical structure, you need to refactor your model to treat hparams as dict.

• strict – Whether to strictly enforce that the keys in checkpoint_path match the keys returned by this module’s state dict.

• **kwargs – Any extra keyword args needed to init the model. Can also be used to override saved hyperparameter values.

Returns

LightningModule instance with loaded weights and hyperparameters (if available).

Note

load_from_checkpoint is a class method. You should use your LightningModule class to call it instead of the LightningModule instance.

Example:

# load weights without mapping ...
model = MyLightningModule.load_from_checkpoint('path/to/checkpoint.ckpt')

# or load weights mapping all weights from GPU 1 to GPU 0 ...
map_location = {'cuda:1':'cuda:0'}
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    map_location=map_location
)

# or load weights and hyperparameters from separate files.
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    hparams_file='/path/to/hparams_file.yaml'
)

# override some of the params with new values
model = MyLightningModule.load_from_checkpoint(
    PATH,
    num_layers=128,
    pretrained_ckpt_path=NEW_PATH,
)

# predict
pretrained_model.eval()
pretrained_model.freeze()
y_hat = pretrained_model(x)

load_state_dict(state_dict: Mapping[str, Any], strict: bool = True)

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Parameters

• state_dict (dict) – a dict containing parameters and persistent buffers.

• strict (bool, optional) – whether to strictly enforce that the keys in state_dict match the keys returned by this module’s state_dict() function. Default: True

Returns

• missing_keys is a list of str containing the missing keys

• unexpected_keys is a list of str containing the unexpected keys

Return type

NamedTuple with missing_keys and unexpected_keys fields

Note

If a parameter or buffer is registered as None and its corresponding key exists in state_dict, load_state_dict() will raise a RuntimeError.

property local_rank: int

The index of the current process within a single node.

log(name: str, value: Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, metric_attribute: Optional[str] = None, rank_zero_only: bool = False) → None

Log a key, value pair.

Example:

self.log('train_loss', loss)

The default behavior per hook is documented here: extensions/logging:Automatic Logging.

Parameters

• name – key to log.

• value – value to log. Can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress bar.

• logger – if True logs to the logger.

• on_step – if True logs at this step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto detach the graph.

• sync_dist – if True, reduces the metric across devices. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the DDP group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple dataloaders). If False, user needs to give unique names for each dataloader to not mix the values.

• batch_size – Current batch_size. This will be directly inferred from the loaded batch, but for some data structures you might need to explicitly provide it.

• metric_attribute – To restore the metric state, Lightning requires the reference of the torchmetrics.Metric in your model. This is found automatically if it is a model attribute.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_dict(dictionary: Mapping[str, Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, rank_zero_only: bool = False) → None

Log a dictionary of values at once.

Example:

values = {'loss': loss, 'acc': acc, ..., 'metric_n': metric_n}
self.log_dict(values)

Parameters

• dictionary – key value pairs. The values can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress base.

• logger – if True logs to the logger.

• on_step – if True logs at this step. None auto-logs for training_step but not validation/test_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. None auto-logs for val/test step but not training_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto-detach the graph

• sync_dist – if True, reduces the metric across GPUs/TPUs. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the ddp group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple). If False, user needs to give unique names for each dataloader to not mix values.

• batch_size – Current batch size. This will be directly inferred from the loaded batch, but some data structures might need to explicitly provide it.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_grad_norm(grad_norm_dict: Dict[str, float]) → None

Override this method to change the default behaviour of log_grad_norm.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

grad_norm_dict – Dictionary containing current grad norm metrics

Example:

# DEFAULT
def log_grad_norm(self, grad_norm_dict):
    self.log_dict(grad_norm_dict, on_step=True, on_epoch=True, prog_bar=False, logger=True)

property logger: Optional[Logger]

Reference to the logger object in the Trainer.

property loggers: List[Logger]

Reference to the list of loggers in the Trainer.

lr_scheduler_step(scheduler: Union[_LRScheduler, ReduceLROnPlateau], optimizer_idx: int, metric: Optional[Any]) → None

Override this method to adjust the default way the Trainer calls each scheduler. By default, Lightning calls step() and as shown in the example for each scheduler based on its interval.

Parameters

• scheduler – Learning rate scheduler.

• optimizer_idx – Index of the optimizer associated with this scheduler.

• metric – Value of the monitor used for schedulers like ReduceLROnPlateau.

Examples:

# DEFAULT
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    if metric is None:
        scheduler.step()
    else:
        scheduler.step(metric)

# Alternative way to update schedulers if it requires an epoch value
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    scheduler.step(epoch=self.current_epoch)

lr_schedulers() → Union[None, List[Union[_LRScheduler, ReduceLROnPlateau]], _LRScheduler, ReduceLROnPlateau]

Returns the learning rate scheduler(s) that are being used during training. Useful for manual optimization.

Returns

A single scheduler, or a list of schedulers in case multiple ones are present, or None if no schedulers were returned in configure_optimizers().

manual_backward(loss: Tensor, *args: Any, **kwargs: Any) → None

Call this directly from your training_step() when doing optimizations manually. By using this, Lightning can ensure that all the proper scaling gets applied when using mixed precision.

See manual optimization for more examples.

Example:

def training_step(...):
    opt = self.optimizers()
    loss = ...
    opt.zero_grad()
    # automatically applies scaling, etc...
    self.manual_backward(loss)
    opt.step()

Parameters

• loss – The tensor on which to compute gradients. Must have a graph attached.

• *args – Additional positional arguments to be forwarded to backward()

• **kwargs – Additional keyword arguments to be forwarded to backward()

modules() → Iterator[Module]

Returns an iterator over all modules in the network.

Yields

Module – a module in the network

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.modules()):
...     print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Tensor]]

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Parameters

• prefix (str) – prefix to prepend to all buffer names.

• recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

(str, torch.Tensor) – Tuple containing the name and buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, buf in self.named_buffers():
>>>    if name in ['running_var']:
>>>        print(buf.size())

named_children() → Iterator[Tuple[str, Module]]

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields

(str, Module) – Tuple containing a name and child module

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, module in model.named_children():
>>>     if name in ['conv4', 'conv5']:
>>>         print(module)

named_modules(memo: Optional[Set[Module]] = None, prefix: str = '', remove_duplicate: bool = True)

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Parameters

• memo – a memo to store the set of modules already added to the result

• prefix – a prefix that will be added to the name of the module

• remove_duplicate – whether to remove the duplicated module instances in the result or not

Yields

(str, Module) – Tuple of name and module

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.named_modules()):
...     print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Parameter]]

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Parameters

• prefix (str) – prefix to prepend to all parameter names.

• recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

(str, Parameter) – Tuple containing the name and parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, param in self.named_parameters():
>>>    if name in ['bias']:
>>>        print(param.size())

on_after_backward() → None

Called after loss.backward() and before optimizers are stepped.

Note

If using native AMP, the gradients will not be unscaled at this point. Use the on_before_optimizer_step if you need the unscaled gradients.

on_after_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch after it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_after_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = gpu_transforms(batch['x'])
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• on_before_batch_transfer()

• transfer_batch_to_device()

on_before_backward(loss: Tensor) → None

Called before loss.backward().

Parameters

loss – Loss divided by number of batches for gradient accumulation and scaled if using native AMP.

on_before_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch before it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_before_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = transforms(batch['x'])
    return batch

See also

• on_after_batch_transfer()

• transfer_batch_to_device()

on_before_optimizer_step(optimizer: Optimizer, optimizer_idx: int) → None

Called before optimizer.step().

If using gradient accumulation, the hook is called once the gradients have been accumulated. See: :paramref:`~pytorch_lightning.trainer.Trainer.accumulate_grad_batches`.

If using native AMP, the loss will be unscaled before calling this hook. See these docs for more information on the scaling of gradients.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

Example:

def on_before_optimizer_step(self, optimizer, optimizer_idx):
    # example to inspect gradient information in tensorboard
    if self.trainer.global_step % 25 == 0:  # don't make the tf file huge
        for k, v in self.named_parameters():
            self.logger.experiment.add_histogram(
                tag=k, values=v.grad, global_step=self.trainer.global_step
            )

on_before_zero_grad(optimizer: Optimizer) → None

Called after training_step() and before optimizer.zero_grad().

Called in the training loop after taking an optimizer step and before zeroing grads. Good place to inspect weight information with weights updated.

This is where it is called:

for optimizer in optimizers:
    out = training_step(...)

    model.on_before_zero_grad(optimizer) # < ---- called here
    optimizer.zero_grad()

    backward()

Parameters

optimizer – The optimizer for which grads should be zeroed.

on_fit_end() → None

Called at the very end of fit.

If on DDP it is called on every process

on_fit_start() → None

Called at the very beginning of fit.

If on DDP it is called on every process

property on_gpu: bool

Returns True if this model is currently located on a GPU.

Useful to set flags around the LightningModule for different CPU vs GPU behavior.

on_load_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning to restore your model. If you saved something with on_save_checkpoint() this is your chance to restore this.

Parameters

checkpoint – Loaded checkpoint

Example:

def on_load_checkpoint(self, checkpoint):
    # 99% of the time you don't need to implement this method
    self.something_cool_i_want_to_save = checkpoint['something_cool_i_want_to_save']

Note

Lightning auto-restores global step, epoch, and train state including amp scaling. There is no need for you to restore anything regarding training.

on_predict_batch_end(outputs: Optional[Any], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop after the batch.

Parameters

• outputs – The outputs of predict_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_end() → None

Called at the end of predicting.

on_predict_epoch_end(results: List[Any]) → None

Called at the end of predicting.

on_predict_epoch_start() → None

Called at the beginning of predicting.

on_predict_model_eval() → None

Sets the model to eval during the predict loop.

on_predict_start() → None

Called at the beginning of predicting.

on_save_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning when saving a checkpoint to give you a chance to store anything else you might want to save.

Parameters

checkpoint – The full checkpoint dictionary before it gets dumped to a file. Implementations of this hook can insert additional data into this dictionary.

Example:

def on_save_checkpoint(self, checkpoint):
    # 99% of use cases you don't need to implement this method
    checkpoint['something_cool_i_want_to_save'] = my_cool_pickable_object

Note

Lightning saves all aspects of training (epoch, global step, etc…) including amp scaling. There is no need for you to store anything about training.

on_test_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop after the batch.

Parameters

• outputs – The outputs of test_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_end() → None

Called at the end of testing.

on_test_epoch_end() → None

Called in the test loop at the very end of the epoch.

on_test_epoch_start() → None

Called in the test loop at the very beginning of the epoch.

on_test_model_eval() → None

Sets the model to eval during the test loop.

on_test_model_train() → None

Sets the model to train during the test loop.

on_test_start() → None

Called at the beginning of testing.

on_train_batch_end(outputs: Union[Tensor, Dict[str, Any]], batch: Any, batch_idx: int) → None

Called in the training loop after the batch.

Parameters

• outputs – The outputs of training_step_end(training_step(x))

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_batch_start(batch: Any, batch_idx: int) → Optional[int]

Called in the training loop before anything happens for that batch.

If you return -1 here, you will skip training for the rest of the current epoch.

Parameters

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_end() → None

Called at the end of training before logger experiment is closed.

on_train_epoch_end() → None

Called in the training loop at the very end of the epoch.

To access all batch outputs at the end of the epoch, either:

1. Implement training_epoch_end in the LightningModule OR

2. Cache data across steps on the attribute(s) of the LightningModule and access them in this hook

on_train_epoch_start() → None

Called in the training loop at the very beginning of the epoch.

on_train_start() → None

Called at the beginning of training after sanity check.

on_validation_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop after the batch.

Parameters

• outputs – The outputs of validation_step_end(validation_step(x))

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_end() → None

Called at the end of validation.

on_validation_epoch_end() → None

Called in the validation loop at the very end of the epoch.

on_validation_epoch_start() → None

Called in the validation loop at the very beginning of the epoch.

on_validation_model_eval() → None

Sets the model to eval during the val loop.

on_validation_model_train() → None

Sets the model to train during the val loop.

on_validation_start() → None

Called at the beginning of validation.

optimizer_step(epoch: int, batch_idx: int, optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int = 0, optimizer_closure: Optional[Callable[[], Any]] = None, on_tpu: bool = False, using_native_amp: bool = False, using_lbfgs: bool = False) → None

Override this method to adjust the default way the Trainer calls each optimizer.

By default, Lightning calls step() and zero_grad() as shown in the example once per optimizer. This method (and zero_grad()) won’t be called during the accumulation phase when Trainer(accumulate_grad_batches != 1). Overriding this hook has no benefit with manual optimization.

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers, this indexes into that list.

• optimizer_closure – The optimizer closure. This closure must be executed as it includes the calls to training_step(), optimizer.zero_grad(), and backward().

• on_tpu – True if TPU backward is required

• using_native_amp – True if using native amp

• using_lbfgs – True if the matching optimizer is torch.optim.LBFGS

Examples:

# DEFAULT
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    optimizer.step(closure=optimizer_closure)

# Alternating schedule for optimizer steps (i.e.: GANs)
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    # update generator opt every step
    if optimizer_idx == 0:
        optimizer.step(closure=optimizer_closure)

    # update discriminator opt every 2 steps
    if optimizer_idx == 1:
        if (batch_idx + 1) % 2 == 0 :
            optimizer.step(closure=optimizer_closure)
        else:
            # call the closure by itself to run `training_step` + `backward` without an optimizer step
            optimizer_closure()

    # ...
    # add as many optimizers as you want

Here’s another example showing how to use this for more advanced things such as learning rate warm-up:

# learning rate warm-up
def optimizer_step(
    self,
    epoch,
    batch_idx,
    optimizer,
    optimizer_idx,
    optimizer_closure,
    on_tpu,
    using_native_amp,
    using_lbfgs,
):
    # update params
    optimizer.step(closure=optimizer_closure)

    # manually warm up lr without a scheduler
    if self.trainer.global_step < 500:
        lr_scale = min(1.0, float(self.trainer.global_step + 1) / 500.0)
        for pg in optimizer.param_groups:
            pg["lr"] = lr_scale * self.learning_rate

optimizer_zero_grad(epoch: int, batch_idx: int, optimizer: Optimizer, optimizer_idx: int) → None

Override this method to change the default behaviour of optimizer.zero_grad().

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers this indexes into that list.

Examples:

# DEFAULT
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad()

# Set gradients to `None` instead of zero to improve performance.
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad(set_to_none=True)

See torch.optim.Optimizer.zero_grad() for the explanation of the above example.

optimizers(use_pl_optimizer: bool = True) → Union[Optimizer, LightningOptimizer, List[Optimizer], List[LightningOptimizer]]

Returns the optimizer(s) that are being used during training. Useful for manual optimization.

Parameters

use_pl_optimizer – If True, will wrap the optimizer(s) in a LightningOptimizer for automatic handling of precision and profiling.

Returns

A single optimizer, or a list of optimizers in case multiple ones are present.

pairwise_correlations(loader: DataLoader, train=False)

Calculates correlation for entire batch from dataloader

Parameters

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

Returns

by default returns the average pairwise correlation in each dimension (for 2 views just the correlation)

parameters(recurse: bool = True) → Iterator[Parameter]

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Parameters

recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

Parameter – module parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for param in model.parameters():
>>>     print(type(param), param.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

predict_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for prediction.

It’s recommended that all data downloads and preparation happen in prepare_data().

• predict()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying prediction samples.

Note

In the case where you return multiple prediction dataloaders, the predict_step() will have an argument dataloader_idx which matches the order here.

predict_step(batch: Any, batch_idx: int, dataloader_idx: int = 0) → Any

Step function called during predict(). By default, it calls forward(). Override to add any processing logic.

The predict_step() is used to scale inference on multi-devices.

To prevent an OOM error, it is possible to use BasePredictionWriter callback to write the predictions to disk or database after each batch or on epoch end.

The BasePredictionWriter should be used while using a spawn based accelerator. This happens for Trainer(strategy="ddp_spawn") or training on 8 TPU cores with Trainer(accelerator="tpu", devices=8) as predictions won’t be returned.

Example

class MyModel(LightningModule):

    def predict_step(self, batch, batch_idx, dataloader_idx=0):
        return self(batch)

dm = ...
model = MyModel()
trainer = Trainer(accelerator="gpu", devices=2)
predictions = trainer.predict(model, dm)

Parameters

• batch – Current batch.

• batch_idx – Index of current batch.

• dataloader_idx – Index of the current dataloader.

Returns

Predicted output

prepare_data() → None

Use this to download and prepare data. Downloading and saving data with multiple processes (distributed settings) will result in corrupted data. Lightning ensures this method is called only within a single process, so you can safely add your downloading logic within.

Warning

DO NOT set state to the model (use setup instead) since this is NOT called on every device

Example:

def prepare_data(self):
    # good
    download_data()
    tokenize()
    etc()

    # bad
    self.split = data_split
    self.some_state = some_other_state()

In a distributed environment, prepare_data can be called in two ways (using prepare_data_per_node)

1. Once per node. This is the default and is only called on LOCAL_RANK=0.

2. Once in total. Only called on GLOBAL_RANK=0.

Example:

# DEFAULT
# called once per node on LOCAL_RANK=0 of that node
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = True


# call on GLOBAL_RANK=0 (great for shared file systems)
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = False

This is called before requesting the dataloaders:

model.prepare_data()
initialize_distributed()
model.setup(stage)
model.train_dataloader()
model.val_dataloader()
model.test_dataloader()
model.predict_dataloader()

print(*args: Any, **kwargs: Any) → None

Prints only from process 0. Use this in any distributed mode to log only once.

Parameters

• *args – The thing to print. The same as for Python’s built-in print function.

• **kwargs – The same as for Python’s built-in print function.

Example:

def forward(self, x):
    self.print(x, 'in forward')

register_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

This function is deprecated in favor of register_full_backward_hook() and the behavior of this function will change in future versions.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_buffer(name: str, tensor: Optional[Tensor], persistent: bool = True) → None

Adds a buffer to the module.

This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the module’s state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module’s state_dict.

Buffers can be accessed as attributes using given names.

Parameters

• name (str) – name of the buffer. The buffer can be accessed from this module using the given name

• tensor (Tensor or None) – buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module’s state_dict.

• persistent (bool) – whether the buffer is part of this module’s state_dict.

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> self.register_buffer('running_mean', torch.zeros(num_features))

register_forward_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output. It should have the following signature:

hook(module, input, output) -> None or modified output

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward() is called.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_forward_pre_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked. It should have the following signature:

hook(module, input) -> None or modified input

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned(unless that value is already a tuple).

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_full_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

The hook will be called every time the gradients with respect to module inputs are computed. The hook should have the following signature:

hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module’s forward function.

Warning

Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_load_state_dict_post_hook(hook)

Registers a post hook to be run after module’s load_state_dict is called.

It should have the following signature::

hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict() with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearning out both missing and unexpected keys will avoid an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_module(name: str, module: Optional[Module]) → None

Alias for add_module().

register_parameter(name: str, param: Optional[Parameter]) → None

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Parameters

• name (str) – name of the parameter. The parameter can be accessed from this module using the given name

• param (Parameter or None) – parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module’s state_dict.

requires_grad_(requires_grad: bool = True) → T

Change if autograd should record operations on parameters in this module.

This method sets the parameters’ requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See Locally disabling gradient computation for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Parameters

requires_grad (bool) – whether autograd should record operations on parameters in this module. Default: True.

Returns

self

Return type

Module

save_hyperparameters(*args: Any, ignore: Optional[Union[Sequence[str], str]] = None, frame: Optional[frame] = None, logger: bool = True) → None

Save arguments to hparams attribute.

Parameters

• args – single object of dict, NameSpace or OmegaConf or string names or arguments from class __init__

• ignore – an argument name or a list of argument names from class __init__ to be ignored

• frame – a frame object. Default is None

• logger – Whether to send the hyperparameters to the logger. Default: True

Example::

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # manually assign arguments
...         self.save_hyperparameters('arg1', 'arg3')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

>>> class AutomaticArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # equivalent automatic
...         self.save_hyperparameters()
...     def forward(self, *args, **kwargs):
...         ...
>>> model = AutomaticArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg2": abc
"arg3": 3.14

>>> class SingleArgModel(HyperparametersMixin):
...     def __init__(self, params):
...         super().__init__()
...         # manually assign single argument
...         self.save_hyperparameters(params)
...     def forward(self, *args, **kwargs):
...         ...
>>> model = SingleArgModel(Namespace(p1=1, p2='abc', p3=3.14))
>>> model.hparams
"p1": 1
"p2": abc
"p3": 3.14

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # pass argument(s) to ignore as a string or in a list
...         self.save_hyperparameters(ignore='arg2')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

score(loader: DataLoader, **kwargs)

Returns average correlation in each dimension (averages over all pairs for multiview)

Parameters

• **kwargs –

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

set_extra_state(state: Any)

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Parameters

state (dict) – Extra state from the state_dict

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters

**params (dict) – Estimator parameters.

Returns

self – Estimator instance.

Return type

estimator instance

setup(stage: str) → None

Called at the beginning of fit (train + validate), validate, test, or predict. This is a good hook when you need to build models dynamically or adjust something about them. This hook is called on every process when using DDP.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

Example:

class LitModel(...):
    def __init__(self):
        self.l1 = None

    def prepare_data(self):
        download_data()
        tokenize()

        # don't do this
        self.something = else

    def setup(self, stage):
        data = load_data(...)
        self.l1 = nn.Linear(28, data.num_classes)

share_memory() → T

See torch.Tensor.share_memory_()

state_dict(*args, destination=None, prefix='', keep_vars=False)

Returns a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

Note

The returned object is a shallow copy. It contains references to the module’s parameters and buffers.

Warning

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

Warning

Please avoid the use of argument destination as it is not designed for end-users.

Parameters

• destination (dict, optional) – If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned. Default: None.

• prefix (str, optional) – a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.

• keep_vars (bool, optional) – by default the Tensor s returned in the state dict are detached from autograd. If it’s set to True, detaching will not be performed. Default: False.

Returns

a dictionary containing a whole state of the module

Return type

dict

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> module.state_dict().keys()
['bias', 'weight']

tbptt_split_batch(batch: Any, split_size: int) → List[Any]

When using truncated backpropagation through time, each batch must be split along the time dimension. Lightning handles this by default, but for custom behavior override this function.

Parameters

• batch – Current batch

• split_size – The size of the split

Returns

List of batch splits. Each split will be passed to training_step() to enable truncated back propagation through time. The default implementation splits root level Tensors and Sequences at dim=1 (i.e. time dim). It assumes that each time dim is the same length.

Examples:

def tbptt_split_batch(self, batch, split_size):
    splits = []
    for t in range(0, time_dims[0], split_size):
        batch_split = []
        for i, x in enumerate(batch):
            if isinstance(x, torch.Tensor):
                split_x = x[:, t:t + split_size]
            elif isinstance(x, collections.abc.Sequence):
                split_x = [None] * len(x)
                for batch_idx in range(len(x)):
                  split_x[batch_idx] = x[batch_idx][t:t + split_size]
            batch_split.append(split_x)
        splits.append(batch_split)
    return splits

Note

Called in the training loop after on_train_batch_start() if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0. Each returned batch split is passed separately to training_step().

teardown(stage: str) → None

Called at the end of fit (train + validate), validate, test, or predict.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

test_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for testing.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• test()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying testing samples.

Example:

def test_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def test_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a test dataset and a test_step(), you don’t need to implement this method.

Note

In the case where you return multiple test dataloaders, the test_step() will have an argument dataloader_idx which matches the order here.

test_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of a test epoch with the output of all test steps.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

outputs – List of outputs you defined in test_step_end(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader

Returns

None

Note

If you didn’t define a test_step(), this won’t be called.

Examples

With a single dataloader:

def test_epoch_end(self, outputs):
    # do something with the outputs of all test batches
    all_test_preds = test_step_outputs.predictions

    some_result = calc_all_results(all_test_preds)
    self.log(some_result)

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each test step for that dataloader.

def test_epoch_end(self, outputs):
    final_value = 0
    for dataloader_outputs in outputs:
        for test_step_out in dataloader_outputs:
            # do something
            final_value += test_step_out

    self.log("final_metric", final_value)

test_step(batch, batch_idx)

Operates on a single batch of data from the test set. In this step you’d normally generate examples or calculate anything of interest such as accuracy.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_id – The index of the dataloader that produced this batch. (only if multiple test dataloaders used).

Returns

Any of.

• Any object or value

• None - Testing will skip to the next batch

# if you have one test dataloader:
def test_step(self, batch, batch_idx):
    ...


# if you have multiple test dataloaders:
def test_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single test dataset
def test_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    test_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'test_loss': loss, 'test_acc': test_acc})

If you pass in multiple test dataloaders, test_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple test dataloaders
def test_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to test you don’t need to implement this method.

Note

When the test_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of the test epoch, the model goes back to training mode and gradients are enabled.

test_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when testing with DP because test_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [test_step(sub_batch) for sub_batch in sub_batches]
test_step_end(step_output)

Parameters

step_output – What you return in test_step() for each batch part.

Returns

None or anything

# WITHOUT test_step_end
# if used in DP, this batch is 1/num_gpus large
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    loss = self.softmax(out)
    self.log("test_loss", loss)


# --------------
# with test_step_end to do softmax over the full batch
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return out


def test_step_end(self, output_results):
    # this out is now the full size of the batch
    all_test_step_outs = output_results.out
    loss = nce_loss(all_test_step_outs)
    self.log("test_loss", loss)

See also

See the Multi GPU Training guide for more details.

to(*args: Any, **kwargs: Any) → Self

See torch.nn.Module.to().

to_empty(*, device: Union[str, device]) → T

Moves the parameters and buffers to the specified device without copying storage.

Parameters

device (torch.device) – The desired device of the parameters and buffers in this module.

Returns

self

Return type

Module

to_onnx(file_path: Union[str, Path], input_sample: Optional[Any] = None, **kwargs: Any) → None

Saves the model in ONNX format.

Parameters

• file_path – The path of the file the onnx model should be saved to.

• input_sample – An input for tracing. Default: None (Use self.example_input_array)

• **kwargs – Will be passed to torch.onnx.export function.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))

>>> with tempfile.NamedTemporaryFile(suffix='.onnx', delete=False) as tmpfile:
...     model = SimpleModel()
...     input_sample = torch.randn((1, 64))
...     model.to_onnx(tmpfile.name, input_sample, export_params=True)
...     os.path.isfile(tmpfile.name)
True

to_torchscript(file_path: Optional[Union[str, Path]] = None, method: Optional[str] = 'script', example_inputs: Optional[Any] = None, **kwargs: Any) → Union[ScriptModule, Dict[str, ScriptModule]]

By default compiles the whole model to a ScriptModule. If you want to use tracing, please provided the argument method='trace' and make sure that either the example_inputs argument is provided, or the model has example_input_array set. If you would like to customize the modules that are scripted you should override this method. In case you want to return multiple modules, we recommend using a dictionary.

Parameters

• file_path – Path where to save the torchscript. Default: None (no file saved).

• method – Whether to use TorchScript’s script or trace method. Default: ‘script’

• example_inputs – An input to be used to do tracing when method is set to ‘trace’. Default: None (uses example_input_array)

• **kwargs – Additional arguments that will be passed to the torch.jit.script() or torch.jit.trace() function.

Note

• Requires the implementation of the forward() method.

• The exported script will be set to evaluation mode.

• It is recommended that you install the latest supported version of PyTorch to use this feature without limitations. See also the torch.jit documentation for supported features.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))
...
>>> model = SimpleModel()
>>> model.to_torchscript(file_path="model.pt")  
>>> os.path.isfile("model.pt")  
>>> torch.jit.save(model.to_torchscript(file_path="model_trace.pt", method='trace', 
...                                     example_inputs=torch.randn(1, 64)))  
>>> os.path.isfile("model_trace.pt")  
True

Returns

This LightningModule as a torchscript, regardless of whether file_path is defined or not.

toggle_optimizer(optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int) → None

Makes sure only the gradients of the current optimizer’s parameters are calculated in the training step to prevent dangling gradients in multiple-optimizer setup.

This is only called automatically when automatic optimization is enabled and multiple optimizers are used. It works with untoggle_optimizer() to make sure param_requires_grad_state is properly reset.

Parameters

• optimizer – The optimizer to toggle.

• optimizer_idx – The index of the optimizer to toggle.

train(mode: bool = True) → T

Sets the module in training mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

Parameters

mode (bool) – whether to set training mode (True) or evaluation mode (False). Default: True.

Returns

self

Return type

Module

train_dataloader() → Union[DataLoader, Sequence[DataLoader], Sequence[Sequence[DataLoader]], Sequence[Dict[str, DataLoader]], Dict[str, DataLoader], Dict[str, Dict[str, DataLoader]], Dict[str, Sequence[DataLoader]]]

Implement one or more PyTorch DataLoaders for training.

Returns

A collection of torch.utils.data.DataLoader specifying training samples. In the case of multiple dataloaders, please see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• fit()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Example:

# single dataloader
def train_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=True, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=True
    )
    return loader

# multiple dataloaders, return as list
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a list of tensors: [batch_mnist, batch_cifar]
    return [mnist_loader, cifar_loader]

# multiple dataloader, return as dict
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a dict of tensors: {'mnist': batch_mnist, 'cifar': batch_cifar}
    return {'mnist': mnist_loader, 'cifar': cifar_loader}

training_epoch_end(outputs: List[Union[Tensor, Dict[str, Any]]]) → None

Called at the end of the training epoch with the outputs of all training steps. Use this in case you need to do something with all the outputs returned by training_step().

# the pseudocode for these calls
train_outs = []
for train_batch in train_data:
    out = training_step(train_batch)
    train_outs.append(out)
training_epoch_end(train_outs)

Parameters

outputs – List of outputs you defined in training_step(). If there are multiple optimizers or when using truncated_bptt_steps > 0, the lists have the dimensions (n_batches, tbptt_steps, n_optimizers). Dimensions of length 1 are squeezed.

Returns

None

Note

If this method is not overridden, this won’t be called.

def training_epoch_end(self, training_step_outputs):
    # do something with all training_step outputs
    for out in training_step_outputs:
        ...

training_step(batch, batch_idx)

Here you compute and return the training loss and some additional metrics for e.g. the progress bar or logger.

Parameters

• batch (Tensor | (Tensor, …) | [Tensor, …]) – The output of your DataLoader. A tensor, tuple or list.

• batch_idx (int) – Integer displaying index of this batch

• optimizer_idx (int) – When using multiple optimizers, this argument will also be present.

• hiddens (Any) – Passed in if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0.

Returns

Any of.

• Tensor - The loss tensor

• dict - A dictionary. Can include any keys, but must include the key 'loss'

• None - Training will skip to the next batch. This is only for automatic optimization.

  This is not supported for multi-GPU, TPU, IPU, or DeepSpeed.

In this step you’d normally do the forward pass and calculate the loss for a batch. You can also do fancier things like multiple forward passes or something model specific.

Example:

def training_step(self, batch, batch_idx):
    x, y, z = batch
    out = self.encoder(x)
    loss = self.loss(out, x)
    return loss

If you define multiple optimizers, this step will be called with an additional optimizer_idx parameter.

# Multiple optimizers (e.g.: GANs)
def training_step(self, batch, batch_idx, optimizer_idx):
    if optimizer_idx == 0:
        # do training_step with encoder
        ...
    if optimizer_idx == 1:
        # do training_step with decoder
        ...

If you add truncated back propagation through time you will also get an additional argument with the hidden states of the previous step.

# Truncated back-propagation through time
def training_step(self, batch, batch_idx, hiddens):
    # hiddens are the hidden states from the previous truncated backprop step
    out, hiddens = self.lstm(data, hiddens)
    loss = ...
    return {"loss": loss, "hiddens": hiddens}

Note

The loss value shown in the progress bar is smoothed (averaged) over the last values, so it differs from the actual loss returned in train/validation step.

Note

When accumulate_grad_batches > 1, the loss returned here will be automatically normalized by accumulate_grad_batches internally.

training_step_end(step_output: Union[Tensor, Dict[str, Any]]) → Union[Tensor, Dict[str, Any]]

Use this when training with dp because training_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [training_step(sub_batch) for sub_batch in sub_batches]
training_step_end(step_output)

Parameters

step_output – What you return in training_step for each batch part.

Returns

Anything

When using the DP strategy, only a portion of the batch is inside the training_step:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)

    # softmax uses only a portion of the batch in the denominator
    loss = self.softmax(out)
    loss = nce_loss(loss)
    return loss

If you wish to do something with all the parts of the batch, then use this method to do it:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return {"pred": out}


def training_step_end(self, training_step_outputs):
    gpu_0_pred = training_step_outputs[0]["pred"]
    gpu_1_pred = training_step_outputs[1]["pred"]
    gpu_n_pred = training_step_outputs[n]["pred"]

    # this softmax now uses the full batch
    loss = nce_loss([gpu_0_pred, gpu_1_pred, gpu_n_pred])
    return loss

See also

See the Multi GPU Training guide for more details.

transfer_batch_to_device(batch: Any, device: device, dataloader_idx: int) → Any

Override this hook if your DataLoader returns tensors wrapped in a custom data structure.

The data types listed below (and any arbitrary nesting of them) are supported out of the box:

• torch.Tensor or anything that implements .to(…)

• list

• dict

• tuple

For anything else, you need to define how the data is moved to the target device (CPU, GPU, TPU, …).

Note

This hook should only transfer the data and not modify it, nor should it move the data to any other device than the one passed in as argument (unless you know what you are doing). To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be transferred to a new device.

• device – The target device as defined in PyTorch.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A reference to the data on the new device.

Example:

def transfer_batch_to_device(self, batch, device, dataloader_idx):
    if isinstance(batch, CustomBatch):
        # move all tensors in your custom data structure to the device
        batch.samples = batch.samples.to(device)
        batch.targets = batch.targets.to(device)
    elif dataloader_idx == 0:
        # skip device transfer for the first dataloader or anything you wish
        pass
    else:
        batch = super().transfer_batch_to_device(data, device, dataloader_idx)
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• move_data_to_device()

• apply_to_collection()

transform(loader: DataLoader)

Parameters

loader – a dataloader that matches the structure of that used for training

Returns

transformed views

property truncated_bptt_steps: int

Enables Truncated Backpropagation Through Time in the Trainer when set to a positive integer.

It represents the number of times training_step() gets called before backpropagation. If this is > 0, the training_step() receives an additional argument hiddens and is expected to return a hidden state.

type(dst_type: Union[str, dtype]) → Self

See torch.nn.Module.type().

unfreeze() → None

Unfreeze all parameters for training.

model = MyLightningModule(...)
model.unfreeze()

untoggle_optimizer(optimizer_idx: int) → None

Resets the state of required gradients that were toggled with toggle_optimizer().

This is only called automatically when automatic optimization is enabled and multiple optimizers are used.

Parameters

optimizer_idx – The index of the optimizer to untoggle.

val_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for validation.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

It’s recommended that all data downloads and preparation happen in prepare_data().

• fit()

• validate()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying validation samples.

Examples:

def val_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False,
                    transform=transform, download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def val_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a validation dataset and a validation_step(), you don’t need to implement this method.

Note

In the case where you return multiple validation dataloaders, the validation_step() will have an argument dataloader_idx which matches the order here.

validation_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of the validation epoch with the outputs of all validation steps.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

outputs – List of outputs you defined in validation_step(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader.

Returns

None

Note

If you didn’t define a validation_step(), this won’t be called.

Examples

With a single dataloader:

def validation_epoch_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each validation step for that dataloader.

def validation_epoch_end(self, outputs):
    for dataloader_output_result in outputs:
        dataloader_outs = dataloader_output_result.dataloader_i_outputs

    self.log("final_metric", final_value)

validation_step(batch, batch_idx)

Operates on a single batch of data from the validation set. In this step you’d might generate examples or calculate anything of interest like accuracy.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_idx – The index of the dataloader that produced this batch. (only if multiple val dataloaders used)

Returns

• Any object or value

• None - Validation will skip to the next batch

# pseudocode of order
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    if defined("validation_step_end"):
        out = validation_step_end(out)
    val_outs.append(out)
val_outs = validation_epoch_end(val_outs)

# if you have one val dataloader:
def validation_step(self, batch, batch_idx):
    ...


# if you have multiple val dataloaders:
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single validation dataset
def validation_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    val_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'val_loss': loss, 'val_acc': val_acc})

If you pass in multiple val dataloaders, validation_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple validation dataloaders
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to validate you don’t need to implement this method.

Note

When the validation_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of validation, the model goes back to training mode and gradients are enabled.

validation_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when validating with dp because validation_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [validation_step(sub_batch) for sub_batch in sub_batches]
validation_step_end(step_output)

Parameters

step_output – What you return in validation_step() for each batch part.

Returns

None or anything

# WITHOUT validation_step_end
# if used in DP, this batch is 1/num_gpus large
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    loss = self.softmax(out)
    loss = nce_loss(loss)
    self.log("val_loss", loss)


# --------------
# with validation_step_end to do softmax over the full batch
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    return out


def validation_step_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

See also

See the Multi GPU Training guide for more details.

xpu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the XPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

zero_grad(set_to_none: bool = False) → None

Sets gradients of all model parameters to zero. See similar function under torch.optim.Optimizer for more context.

Parameters

set_to_none (bool) – instead of setting to zero, set the grads to None. See torch.optim.Optimizer.zero_grad() for details.

class cca_zoo.deepmodels.DCCA_SDL(latent_dims: int, N: int, encoders=None, r: float = 0, rho: float = 0.2, eps: float = 1e-05, shared_target: bool = False, lam=0.5, **kwargs)

Bases: DCCA_NOI

A class used to fit a Deep CCA by Stochastic Decorrelation model.

References

Chang, Xiaobin, Tao Xiang, and Timothy M. Hospedales. “Scalable and effective deep CCA via soft decorrelation.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018.

forward(views, **kwargs)

We use the forward model to define the transformation of views to the latent space :param views: batches for each view separated by commas

loss(views, **kwargs)

Loss for model :param views: :param args: :param kwargs: :return:

add_module(name: str, module: Optional[Module]) → None

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Parameters

• name (str) – name of the child module. The child module can be accessed from this module using the given name

• module (Module) – child module to be added to the module.

all_gather(data: Union[Tensor, Dict, List, Tuple], group: Optional[Any] = None, sync_grads: bool = False) → Union[Tensor, Dict, List, Tuple]

Allows users to call self.all_gather() from the LightningModule, thus making the all_gather operation accelerator agnostic. all_gather is a function provided by accelerators to gather a tensor from several distributed processes.

Parameters

• data – int, float, tensor of shape (batch, …), or a (possibly nested) collection thereof.

• group – the process group to gather results from. Defaults to all processes (world)

• sync_grads – flag that allows users to synchronize gradients for the all_gather operation

Returns

A tensor of shape (world_size, batch, …), or if the input was a collection the output will also be a collection with tensors of this shape.

apply(fn: Callable[[Module], None]) → T

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also torch.nn.init).

Parameters

fn (Module -> None) – function to be applied to each submodule

Returns

self

Return type

Module

Example:

>>> @torch.no_grad()
>>> def init_weights(m):
>>>     print(m)
>>>     if type(m) == nn.Linear:
>>>         m.weight.fill_(1.0)
>>>         print(m.weight)
>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

property automatic_optimization: bool

If set to False you are responsible for calling .backward(), .step(), .zero_grad().

backward(loss: Tensor, optimizer: Optional[Steppable], optimizer_idx: Optional[int], *args: Any, **kwargs: Any) → None

Called to perform backward on the loss returned in training_step(). Override this hook with your own implementation if you need to.

Parameters

• loss – The loss tensor returned by training_step(). If gradient accumulation is used, the loss here holds the normalized value (scaled by 1 / accumulation steps).

• optimizer – Current optimizer being used. None if using manual optimization.

• optimizer_idx – Index of the current optimizer being used. None if using manual optimization.

Example:

def backward(self, loss, optimizer, optimizer_idx):
    loss.backward()

bfloat16() → T

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns

self

Return type

Module

buffers(recurse: bool = True) → Iterator[Tensor]

Returns an iterator over module buffers.

Parameters

recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

torch.Tensor – module buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for buf in model.buffers():
>>>     print(type(buf), buf.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

children() → Iterator[Module]

Returns an iterator over immediate children modules.

Yields

Module – a child module

clip_gradients(optimizer: Optimizer, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Handles gradient clipping internally.

Note

Do not override this method. If you want to customize gradient clipping, consider using configure_gradient_clipping() method.

Parameters

• optimizer – Current optimizer being used.

• gradient_clip_val – The value at which to clip gradients.

• gradient_clip_algorithm – The gradient clipping algorithm to use. Pass gradient_clip_algorithm="value" to clip by value, and gradient_clip_algorithm="norm" to clip by norm.

configure_callbacks()

Configure model-specific callbacks. When the model gets attached, e.g., when .fit() or .test() gets called, the list or a callback returned here will be merged with the list of callbacks passed to the Trainer’s callbacks argument. If a callback returned here has the same type as one or several callbacks already present in the Trainer’s callbacks list, it will take priority and replace them. In addition, Lightning will make sure ModelCheckpoint callbacks run last.

Returns

A callback or a list of callbacks which will extend the list of callbacks in the Trainer.

Example:

def configure_callbacks(self):
    early_stop = EarlyStopping(monitor="val_acc", mode="max")
    checkpoint = ModelCheckpoint(monitor="val_loss")
    return [early_stop, checkpoint]

configure_gradient_clipping(optimizer: Optimizer, optimizer_idx: int, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Perform gradient clipping for the optimizer parameters. Called before optimizer_step().

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

• gradient_clip_val – The value at which to clip gradients. By default value passed in Trainer will be available here.

• gradient_clip_algorithm – The gradient clipping algorithm to use. By default value passed in Trainer will be available here.

Example:

# Perform gradient clipping on gradients associated with discriminator (optimizer_idx=1) in GAN
def configure_gradient_clipping(self, optimizer, optimizer_idx, gradient_clip_val, gradient_clip_algorithm):
    if optimizer_idx == 1:
        # Lightning will handle the gradient clipping
        self.clip_gradients(
            optimizer,
            gradient_clip_val=gradient_clip_val,
            gradient_clip_algorithm=gradient_clip_algorithm
        )
    else:
        # implement your own custom logic to clip gradients for generator (optimizer_idx=0)

configure_optimizers()

Collects learnable parameters and configures the optimizer and learning rate scheduler. :returns: two lists containing the optimizer and the scheduler. :rtype: Tuple[List, List]

configure_sharded_model() → None

Hook to create modules in a distributed aware context. This is useful for when using sharded plugins, where we’d like to shard the model instantly, which is useful for extremely large models which can save memory and initialization time.

This hook is called during each of fit/val/test/predict stages in the same process, so ensure that implementation of this hook is idempotent.

cpu() → Self

See torch.nn.Module.cpu().

cuda(device: Optional[Union[device, int]] = None) → Self

Moves all model parameters and buffers to the GPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized.

Parameters

device – If specified, all parameters will be copied to that device. If None, the current CUDA device index will be used.

Returns

self

Return type

Module

property current_epoch: int

The current epoch in the Trainer, or 0 if not attached.

double() → Self

See torch.nn.Module.double().

eval() → T

Sets the module in evaluation mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False).

See Locally disabling gradient computation for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns

self

Return type

Module

property example_input_array: Optional[Union[Tensor, Tuple, Dict]]

The example input array is a specification of what the module can consume in the forward() method. The return type is interpreted as follows:

• Single tensor: It is assumed the model takes a single argument, i.e., model.forward(model.example_input_array)

• Tuple: The input array should be interpreted as a sequence of positional arguments, i.e., model.forward(*model.example_input_array)

• Dict: The input array represents named keyword arguments, i.e., model.forward(**model.example_input_array)

extra_repr() → str

Set the extra representation of the module

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

abstract fit(views: Iterable[ndarray], y=None, **kwargs)

Fits the model to the given data

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• y (None) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

self

Return type

object

fit_transform(views: Iterable[ndarray], **kwargs)

Fits the model to the given data and returns the transformed views

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

transformed_views

Return type

list of numpy arrays

float() → Self

See torch.nn.Module.float().

freeze() → None

Freeze all params for inference.

Example:

model = MyLightningModule(...)
model.freeze()

get_buffer(target: str) → Tensor

Returns the buffer given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The buffer referenced by target

Return type

torch.Tensor

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state() → Any

Returns any extra state to include in the module’s state_dict. Implement this and a corresponding set_extra_state() for your module if you need to store extra state. This function is called when building the module’s state_dict().

Note that extra state should be pickleable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns

Any extra state to store in the module’s state_dict

Return type

object

get_factor_loadings(views: Iterable[ndarray], normalize=True, **kwargs)

Returns the factor loadings for each view

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• normalize (bool, optional) – Whether to normalize the factor loadings. Default is True.

• kwargs (any additional keyword arguments required by the given model) –

Returns

factor_loadings

Return type

list of numpy arrays

get_parameter(target: str) → Parameter

Returns the parameter given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The Parameter referenced by target

Return type

torch.nn.Parameter

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_params(deep=True)

Get parameters for this estimator.

Parameters

deep (bool, default=True) – If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns

params – Parameter names mapped to their values.

Return type

dict

get_submodule(target: str) → Module

Returns the submodule given by target if it exists, otherwise throws an error.

For example, let’s say you have an nn.Module A that looks like this:

A(
    (net_b): Module(
        (net_c): Module(
            (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2))
        )
        (linear): Linear(in_features=100, out_features=200, bias=True)
    )
)

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Parameters

target – The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns

The submodule referenced by target

Return type

torch.nn.Module

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Module

property global_rank: int

The index of the current process across all nodes and devices.

property global_step: int

Total training batches seen across all epochs.

If no Trainer is attached, this propery is 0.

half() → Self

See torch.nn.Module.half().

property hparams: Union[AttributeDict, MutableMapping]

The collection of hyperparameters saved with save_hyperparameters(). It is mutable by the user. For the frozen set of initial hyperparameters, use hparams_initial.

Returns

Mutable hyperparameters dictionary

property hparams_initial: AttributeDict

The collection of hyperparameters saved with save_hyperparameters(). These contents are read-only. Manual updates to the saved hyperparameters can instead be performed through hparams.

Returns

immutable initial hyperparameters

Return type

AttributeDict

ipu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the IPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on IPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

classmethod load_from_checkpoint(checkpoint_path: Union[str, Path, IO], map_location: Optional[Union[device, str, int, Callable[[Union[device, str, int]], Union[device, str, int]], Dict[Union[device, str, int], Union[device, str, int]]]] = None, hparams_file: Optional[Union[str, Path]] = None, strict: bool = True, **kwargs: Any) → Self

Primary way of loading a model from a checkpoint. When Lightning saves a checkpoint it stores the arguments passed to __init__ in the checkpoint under "hyper_parameters".

Any arguments specified through **kwargs will override args stored in "hyper_parameters".

Parameters

• checkpoint_path – Path to checkpoint. This can also be a URL, or file-like object

• map_location – If your checkpoint saved a GPU model and you now load on CPUs or a different number of GPUs, use this to map to the new setup. The behaviour is the same as in torch.load().

• hparams_file –

  Optional path to a .yaml or .csv file with hierarchical structure as in this example:

  drop_prob: 0.2
  dataloader:
      batch_size: 32
  

  You most likely won’t need this since Lightning will always save the hyperparameters to the checkpoint. However, if your checkpoint weights don’t have the hyperparameters saved, use this method to pass in a .yaml file with the hparams you’d like to use. These will be converted into a dict and passed into your LightningModule for use.

  If your model’s hparams argument is Namespace and .yaml file has hierarchical structure, you need to refactor your model to treat hparams as dict.

• strict – Whether to strictly enforce that the keys in checkpoint_path match the keys returned by this module’s state dict.

• **kwargs – Any extra keyword args needed to init the model. Can also be used to override saved hyperparameter values.

Returns

LightningModule instance with loaded weights and hyperparameters (if available).

Note

load_from_checkpoint is a class method. You should use your LightningModule class to call it instead of the LightningModule instance.

Example:

# load weights without mapping ...
model = MyLightningModule.load_from_checkpoint('path/to/checkpoint.ckpt')

# or load weights mapping all weights from GPU 1 to GPU 0 ...
map_location = {'cuda:1':'cuda:0'}
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    map_location=map_location
)

# or load weights and hyperparameters from separate files.
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    hparams_file='/path/to/hparams_file.yaml'
)

# override some of the params with new values
model = MyLightningModule.load_from_checkpoint(
    PATH,
    num_layers=128,
    pretrained_ckpt_path=NEW_PATH,
)

# predict
pretrained_model.eval()
pretrained_model.freeze()
y_hat = pretrained_model(x)

load_state_dict(state_dict: Mapping[str, Any], strict: bool = True)

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Parameters

• state_dict (dict) – a dict containing parameters and persistent buffers.

• strict (bool, optional) – whether to strictly enforce that the keys in state_dict match the keys returned by this module’s state_dict() function. Default: True

Returns

• missing_keys is a list of str containing the missing keys

• unexpected_keys is a list of str containing the unexpected keys

Return type

NamedTuple with missing_keys and unexpected_keys fields

Note

If a parameter or buffer is registered as None and its corresponding key exists in state_dict, load_state_dict() will raise a RuntimeError.

property local_rank: int

The index of the current process within a single node.

log(name: str, value: Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, metric_attribute: Optional[str] = None, rank_zero_only: bool = False) → None

Log a key, value pair.

Example:

self.log('train_loss', loss)

The default behavior per hook is documented here: extensions/logging:Automatic Logging.

Parameters

• name – key to log.

• value – value to log. Can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress bar.

• logger – if True logs to the logger.

• on_step – if True logs at this step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto detach the graph.

• sync_dist – if True, reduces the metric across devices. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the DDP group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple dataloaders). If False, user needs to give unique names for each dataloader to not mix the values.

• batch_size – Current batch_size. This will be directly inferred from the loaded batch, but for some data structures you might need to explicitly provide it.

• metric_attribute – To restore the metric state, Lightning requires the reference of the torchmetrics.Metric in your model. This is found automatically if it is a model attribute.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_dict(dictionary: Mapping[str, Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, rank_zero_only: bool = False) → None

Log a dictionary of values at once.

Example:

values = {'loss': loss, 'acc': acc, ..., 'metric_n': metric_n}
self.log_dict(values)

Parameters

• dictionary – key value pairs. The values can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress base.

• logger – if True logs to the logger.

• on_step – if True logs at this step. None auto-logs for training_step but not validation/test_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. None auto-logs for val/test step but not training_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto-detach the graph

• sync_dist – if True, reduces the metric across GPUs/TPUs. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the ddp group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple). If False, user needs to give unique names for each dataloader to not mix values.

• batch_size – Current batch size. This will be directly inferred from the loaded batch, but some data structures might need to explicitly provide it.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_grad_norm(grad_norm_dict: Dict[str, float]) → None

Override this method to change the default behaviour of log_grad_norm.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

grad_norm_dict – Dictionary containing current grad norm metrics

Example:

# DEFAULT
def log_grad_norm(self, grad_norm_dict):
    self.log_dict(grad_norm_dict, on_step=True, on_epoch=True, prog_bar=False, logger=True)

property logger: Optional[Logger]

Reference to the logger object in the Trainer.

property loggers: List[Logger]

Reference to the list of loggers in the Trainer.

lr_scheduler_step(scheduler: Union[_LRScheduler, ReduceLROnPlateau], optimizer_idx: int, metric: Optional[Any]) → None

Override this method to adjust the default way the Trainer calls each scheduler. By default, Lightning calls step() and as shown in the example for each scheduler based on its interval.

Parameters

• scheduler – Learning rate scheduler.

• optimizer_idx – Index of the optimizer associated with this scheduler.

• metric – Value of the monitor used for schedulers like ReduceLROnPlateau.

Examples:

# DEFAULT
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    if metric is None:
        scheduler.step()
    else:
        scheduler.step(metric)

# Alternative way to update schedulers if it requires an epoch value
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    scheduler.step(epoch=self.current_epoch)

lr_schedulers() → Union[None, List[Union[_LRScheduler, ReduceLROnPlateau]], _LRScheduler, ReduceLROnPlateau]

Returns the learning rate scheduler(s) that are being used during training. Useful for manual optimization.

Returns

A single scheduler, or a list of schedulers in case multiple ones are present, or None if no schedulers were returned in configure_optimizers().

manual_backward(loss: Tensor, *args: Any, **kwargs: Any) → None

Call this directly from your training_step() when doing optimizations manually. By using this, Lightning can ensure that all the proper scaling gets applied when using mixed precision.

See manual optimization for more examples.

Example:

def training_step(...):
    opt = self.optimizers()
    loss = ...
    opt.zero_grad()
    # automatically applies scaling, etc...
    self.manual_backward(loss)
    opt.step()

Parameters

• loss – The tensor on which to compute gradients. Must have a graph attached.

• *args – Additional positional arguments to be forwarded to backward()

• **kwargs – Additional keyword arguments to be forwarded to backward()

modules() → Iterator[Module]

Returns an iterator over all modules in the network.

Yields

Module – a module in the network

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.modules()):
...     print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Tensor]]

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Parameters

• prefix (str) – prefix to prepend to all buffer names.

• recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

(str, torch.Tensor) – Tuple containing the name and buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, buf in self.named_buffers():
>>>    if name in ['running_var']:
>>>        print(buf.size())

named_children() → Iterator[Tuple[str, Module]]

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields

(str, Module) – Tuple containing a name and child module

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, module in model.named_children():
>>>     if name in ['conv4', 'conv5']:
>>>         print(module)

named_modules(memo: Optional[Set[Module]] = None, prefix: str = '', remove_duplicate: bool = True)

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Parameters

• memo – a memo to store the set of modules already added to the result

• prefix – a prefix that will be added to the name of the module

• remove_duplicate – whether to remove the duplicated module instances in the result or not

Yields

(str, Module) – Tuple of name and module

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.named_modules()):
...     print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Parameter]]

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Parameters

• prefix (str) – prefix to prepend to all parameter names.

• recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

(str, Parameter) – Tuple containing the name and parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, param in self.named_parameters():
>>>    if name in ['bias']:
>>>        print(param.size())

on_after_backward() → None

Called after loss.backward() and before optimizers are stepped.

Note

If using native AMP, the gradients will not be unscaled at this point. Use the on_before_optimizer_step if you need the unscaled gradients.

on_after_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch after it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_after_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = gpu_transforms(batch['x'])
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• on_before_batch_transfer()

• transfer_batch_to_device()

on_before_backward(loss: Tensor) → None

Called before loss.backward().

Parameters

loss – Loss divided by number of batches for gradient accumulation and scaled if using native AMP.

on_before_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch before it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_before_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = transforms(batch['x'])
    return batch

See also

• on_after_batch_transfer()

• transfer_batch_to_device()

on_before_optimizer_step(optimizer: Optimizer, optimizer_idx: int) → None

Called before optimizer.step().

If using gradient accumulation, the hook is called once the gradients have been accumulated. See: :paramref:`~pytorch_lightning.trainer.Trainer.accumulate_grad_batches`.

If using native AMP, the loss will be unscaled before calling this hook. See these docs for more information on the scaling of gradients.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

Example:

def on_before_optimizer_step(self, optimizer, optimizer_idx):
    # example to inspect gradient information in tensorboard
    if self.trainer.global_step % 25 == 0:  # don't make the tf file huge
        for k, v in self.named_parameters():
            self.logger.experiment.add_histogram(
                tag=k, values=v.grad, global_step=self.trainer.global_step
            )

on_before_zero_grad(optimizer: Optimizer) → None

Called after training_step() and before optimizer.zero_grad().

Called in the training loop after taking an optimizer step and before zeroing grads. Good place to inspect weight information with weights updated.

This is where it is called:

for optimizer in optimizers:
    out = training_step(...)

    model.on_before_zero_grad(optimizer) # < ---- called here
    optimizer.zero_grad()

    backward()

Parameters

optimizer – The optimizer for which grads should be zeroed.

on_fit_end() → None

Called at the very end of fit.

If on DDP it is called on every process

on_fit_start() → None

Called at the very beginning of fit.

If on DDP it is called on every process

property on_gpu: bool

Returns True if this model is currently located on a GPU.

Useful to set flags around the LightningModule for different CPU vs GPU behavior.

on_load_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning to restore your model. If you saved something with on_save_checkpoint() this is your chance to restore this.

Parameters

checkpoint – Loaded checkpoint

Example:

def on_load_checkpoint(self, checkpoint):
    # 99% of the time you don't need to implement this method
    self.something_cool_i_want_to_save = checkpoint['something_cool_i_want_to_save']

Note

Lightning auto-restores global step, epoch, and train state including amp scaling. There is no need for you to restore anything regarding training.

on_predict_batch_end(outputs: Optional[Any], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop after the batch.

Parameters

• outputs – The outputs of predict_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_end() → None

Called at the end of predicting.

on_predict_epoch_end(results: List[Any]) → None

Called at the end of predicting.

on_predict_epoch_start() → None

Called at the beginning of predicting.

on_predict_model_eval() → None

Sets the model to eval during the predict loop.

on_predict_start() → None

Called at the beginning of predicting.

on_save_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning when saving a checkpoint to give you a chance to store anything else you might want to save.

Parameters

checkpoint – The full checkpoint dictionary before it gets dumped to a file. Implementations of this hook can insert additional data into this dictionary.

Example:

def on_save_checkpoint(self, checkpoint):
    # 99% of use cases you don't need to implement this method
    checkpoint['something_cool_i_want_to_save'] = my_cool_pickable_object

Note

Lightning saves all aspects of training (epoch, global step, etc…) including amp scaling. There is no need for you to store anything about training.

on_test_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop after the batch.

Parameters

• outputs – The outputs of test_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_end() → None

Called at the end of testing.

on_test_epoch_end() → None

Called in the test loop at the very end of the epoch.

on_test_epoch_start() → None

Called in the test loop at the very beginning of the epoch.

on_test_model_eval() → None

Sets the model to eval during the test loop.

on_test_model_train() → None

Sets the model to train during the test loop.

on_test_start() → None

Called at the beginning of testing.

on_train_batch_end(outputs: Union[Tensor, Dict[str, Any]], batch: Any, batch_idx: int) → None

Called in the training loop after the batch.

Parameters

• outputs – The outputs of training_step_end(training_step(x))

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_batch_start(batch: Any, batch_idx: int) → Optional[int]

Called in the training loop before anything happens for that batch.

If you return -1 here, you will skip training for the rest of the current epoch.

Parameters

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_end() → None

Called at the end of training before logger experiment is closed.

on_train_epoch_end() → None

Called in the training loop at the very end of the epoch.

To access all batch outputs at the end of the epoch, either:

1. Implement training_epoch_end in the LightningModule OR

2. Cache data across steps on the attribute(s) of the LightningModule and access them in this hook

on_train_epoch_start() → None

Called in the training loop at the very beginning of the epoch.

on_train_start() → None

Called at the beginning of training after sanity check.

on_validation_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop after the batch.

Parameters

• outputs – The outputs of validation_step_end(validation_step(x))

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_end() → None

Called at the end of validation.

on_validation_epoch_end() → None

Called in the validation loop at the very end of the epoch.

on_validation_epoch_start() → None

Called in the validation loop at the very beginning of the epoch.

on_validation_model_eval() → None

Sets the model to eval during the val loop.

on_validation_model_train() → None

Sets the model to train during the val loop.

on_validation_start() → None

Called at the beginning of validation.

optimizer_step(epoch: int, batch_idx: int, optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int = 0, optimizer_closure: Optional[Callable[[], Any]] = None, on_tpu: bool = False, using_native_amp: bool = False, using_lbfgs: bool = False) → None

Override this method to adjust the default way the Trainer calls each optimizer.

By default, Lightning calls step() and zero_grad() as shown in the example once per optimizer. This method (and zero_grad()) won’t be called during the accumulation phase when Trainer(accumulate_grad_batches != 1). Overriding this hook has no benefit with manual optimization.

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers, this indexes into that list.

• optimizer_closure – The optimizer closure. This closure must be executed as it includes the calls to training_step(), optimizer.zero_grad(), and backward().

• on_tpu – True if TPU backward is required

• using_native_amp – True if using native amp

• using_lbfgs – True if the matching optimizer is torch.optim.LBFGS

Examples:

# DEFAULT
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    optimizer.step(closure=optimizer_closure)

# Alternating schedule for optimizer steps (i.e.: GANs)
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    # update generator opt every step
    if optimizer_idx == 0:
        optimizer.step(closure=optimizer_closure)

    # update discriminator opt every 2 steps
    if optimizer_idx == 1:
        if (batch_idx + 1) % 2 == 0 :
            optimizer.step(closure=optimizer_closure)
        else:
            # call the closure by itself to run `training_step` + `backward` without an optimizer step
            optimizer_closure()

    # ...
    # add as many optimizers as you want

Here’s another example showing how to use this for more advanced things such as learning rate warm-up:

# learning rate warm-up
def optimizer_step(
    self,
    epoch,
    batch_idx,
    optimizer,
    optimizer_idx,
    optimizer_closure,
    on_tpu,
    using_native_amp,
    using_lbfgs,
):
    # update params
    optimizer.step(closure=optimizer_closure)

    # manually warm up lr without a scheduler
    if self.trainer.global_step < 500:
        lr_scale = min(1.0, float(self.trainer.global_step + 1) / 500.0)
        for pg in optimizer.param_groups:
            pg["lr"] = lr_scale * self.learning_rate

optimizer_zero_grad(epoch: int, batch_idx: int, optimizer: Optimizer, optimizer_idx: int) → None

Override this method to change the default behaviour of optimizer.zero_grad().

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers this indexes into that list.

Examples:

# DEFAULT
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad()

# Set gradients to `None` instead of zero to improve performance.
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad(set_to_none=True)

See torch.optim.Optimizer.zero_grad() for the explanation of the above example.

optimizers(use_pl_optimizer: bool = True) → Union[Optimizer, LightningOptimizer, List[Optimizer], List[LightningOptimizer]]

Returns the optimizer(s) that are being used during training. Useful for manual optimization.

Parameters

use_pl_optimizer – If True, will wrap the optimizer(s) in a LightningOptimizer for automatic handling of precision and profiling.

Returns

A single optimizer, or a list of optimizers in case multiple ones are present.

pairwise_correlations(loader: DataLoader, train=False)

Calculates correlation for entire batch from dataloader

Parameters

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

Returns

by default returns the average pairwise correlation in each dimension (for 2 views just the correlation)

parameters(recurse: bool = True) → Iterator[Parameter]

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Parameters

recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

Parameter – module parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for param in model.parameters():
>>>     print(type(param), param.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

predict_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for prediction.

It’s recommended that all data downloads and preparation happen in prepare_data().

• predict()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying prediction samples.

Note

In the case where you return multiple prediction dataloaders, the predict_step() will have an argument dataloader_idx which matches the order here.

predict_step(batch: Any, batch_idx: int, dataloader_idx: int = 0) → Any

Step function called during predict(). By default, it calls forward(). Override to add any processing logic.

The predict_step() is used to scale inference on multi-devices.

To prevent an OOM error, it is possible to use BasePredictionWriter callback to write the predictions to disk or database after each batch or on epoch end.

The BasePredictionWriter should be used while using a spawn based accelerator. This happens for Trainer(strategy="ddp_spawn") or training on 8 TPU cores with Trainer(accelerator="tpu", devices=8) as predictions won’t be returned.

Example

class MyModel(LightningModule):

    def predict_step(self, batch, batch_idx, dataloader_idx=0):
        return self(batch)

dm = ...
model = MyModel()
trainer = Trainer(accelerator="gpu", devices=2)
predictions = trainer.predict(model, dm)

Parameters

• batch – Current batch.

• batch_idx – Index of current batch.

• dataloader_idx – Index of the current dataloader.

Returns

Predicted output

prepare_data() → None

Use this to download and prepare data. Downloading and saving data with multiple processes (distributed settings) will result in corrupted data. Lightning ensures this method is called only within a single process, so you can safely add your downloading logic within.

Warning

DO NOT set state to the model (use setup instead) since this is NOT called on every device

Example:

def prepare_data(self):
    # good
    download_data()
    tokenize()
    etc()

    # bad
    self.split = data_split
    self.some_state = some_other_state()

In a distributed environment, prepare_data can be called in two ways (using prepare_data_per_node)

1. Once per node. This is the default and is only called on LOCAL_RANK=0.

2. Once in total. Only called on GLOBAL_RANK=0.

Example:

# DEFAULT
# called once per node on LOCAL_RANK=0 of that node
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = True


# call on GLOBAL_RANK=0 (great for shared file systems)
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = False

This is called before requesting the dataloaders:

model.prepare_data()
initialize_distributed()
model.setup(stage)
model.train_dataloader()
model.val_dataloader()
model.test_dataloader()
model.predict_dataloader()

print(*args: Any, **kwargs: Any) → None

Prints only from process 0. Use this in any distributed mode to log only once.

Parameters

• *args – The thing to print. The same as for Python’s built-in print function.

• **kwargs – The same as for Python’s built-in print function.

Example:

def forward(self, x):
    self.print(x, 'in forward')

register_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

This function is deprecated in favor of register_full_backward_hook() and the behavior of this function will change in future versions.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_buffer(name: str, tensor: Optional[Tensor], persistent: bool = True) → None

Adds a buffer to the module.

This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the module’s state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module’s state_dict.

Buffers can be accessed as attributes using given names.

Parameters

• name (str) – name of the buffer. The buffer can be accessed from this module using the given name

• tensor (Tensor or None) – buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module’s state_dict.

• persistent (bool) – whether the buffer is part of this module’s state_dict.

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> self.register_buffer('running_mean', torch.zeros(num_features))

register_forward_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output. It should have the following signature:

hook(module, input, output) -> None or modified output

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward() is called.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_forward_pre_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked. It should have the following signature:

hook(module, input) -> None or modified input

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned(unless that value is already a tuple).

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_full_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

The hook will be called every time the gradients with respect to module inputs are computed. The hook should have the following signature:

hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module’s forward function.

Warning

Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_load_state_dict_post_hook(hook)

Registers a post hook to be run after module’s load_state_dict is called.

It should have the following signature::

hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict() with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearning out both missing and unexpected keys will avoid an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_module(name: str, module: Optional[Module]) → None

Alias for add_module().

register_parameter(name: str, param: Optional[Parameter]) → None

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Parameters

• name (str) – name of the parameter. The parameter can be accessed from this module using the given name

• param (Parameter or None) – parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module’s state_dict.

requires_grad_(requires_grad: bool = True) → T

Change if autograd should record operations on parameters in this module.

This method sets the parameters’ requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See Locally disabling gradient computation for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Parameters

requires_grad (bool) – whether autograd should record operations on parameters in this module. Default: True.

Returns

self

Return type

Module

save_hyperparameters(*args: Any, ignore: Optional[Union[Sequence[str], str]] = None, frame: Optional[frame] = None, logger: bool = True) → None

Save arguments to hparams attribute.

Parameters

• args – single object of dict, NameSpace or OmegaConf or string names or arguments from class __init__

• ignore – an argument name or a list of argument names from class __init__ to be ignored

• frame – a frame object. Default is None

• logger – Whether to send the hyperparameters to the logger. Default: True

Example::

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # manually assign arguments
...         self.save_hyperparameters('arg1', 'arg3')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

>>> class AutomaticArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # equivalent automatic
...         self.save_hyperparameters()
...     def forward(self, *args, **kwargs):
...         ...
>>> model = AutomaticArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg2": abc
"arg3": 3.14

>>> class SingleArgModel(HyperparametersMixin):
...     def __init__(self, params):
...         super().__init__()
...         # manually assign single argument
...         self.save_hyperparameters(params)
...     def forward(self, *args, **kwargs):
...         ...
>>> model = SingleArgModel(Namespace(p1=1, p2='abc', p3=3.14))
>>> model.hparams
"p1": 1
"p2": abc
"p3": 3.14

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # pass argument(s) to ignore as a string or in a list
...         self.save_hyperparameters(ignore='arg2')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

score(loader: DataLoader, **kwargs)

Returns average correlation in each dimension (averages over all pairs for multiview)

Parameters

• **kwargs –

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

set_extra_state(state: Any)

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Parameters

state (dict) – Extra state from the state_dict

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters

**params (dict) – Estimator parameters.

Returns

self – Estimator instance.

Return type

estimator instance

setup(stage: str) → None

Called at the beginning of fit (train + validate), validate, test, or predict. This is a good hook when you need to build models dynamically or adjust something about them. This hook is called on every process when using DDP.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

Example:

class LitModel(...):
    def __init__(self):
        self.l1 = None

    def prepare_data(self):
        download_data()
        tokenize()

        # don't do this
        self.something = else

    def setup(self, stage):
        data = load_data(...)
        self.l1 = nn.Linear(28, data.num_classes)

share_memory() → T

See torch.Tensor.share_memory_()

state_dict(*args, destination=None, prefix='', keep_vars=False)

Returns a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

Note

The returned object is a shallow copy. It contains references to the module’s parameters and buffers.

Warning

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

Warning

Please avoid the use of argument destination as it is not designed for end-users.

Parameters

• destination (dict, optional) – If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned. Default: None.

• prefix (str, optional) – a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.

• keep_vars (bool, optional) – by default the Tensor s returned in the state dict are detached from autograd. If it’s set to True, detaching will not be performed. Default: False.

Returns

a dictionary containing a whole state of the module

Return type

dict

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> module.state_dict().keys()
['bias', 'weight']

tbptt_split_batch(batch: Any, split_size: int) → List[Any]

When using truncated backpropagation through time, each batch must be split along the time dimension. Lightning handles this by default, but for custom behavior override this function.

Parameters

• batch – Current batch

• split_size – The size of the split

Returns

List of batch splits. Each split will be passed to training_step() to enable truncated back propagation through time. The default implementation splits root level Tensors and Sequences at dim=1 (i.e. time dim). It assumes that each time dim is the same length.

Examples:

def tbptt_split_batch(self, batch, split_size):
    splits = []
    for t in range(0, time_dims[0], split_size):
        batch_split = []
        for i, x in enumerate(batch):
            if isinstance(x, torch.Tensor):
                split_x = x[:, t:t + split_size]
            elif isinstance(x, collections.abc.Sequence):
                split_x = [None] * len(x)
                for batch_idx in range(len(x)):
                  split_x[batch_idx] = x[batch_idx][t:t + split_size]
            batch_split.append(split_x)
        splits.append(batch_split)
    return splits

Note

Called in the training loop after on_train_batch_start() if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0. Each returned batch split is passed separately to training_step().

teardown(stage: str) → None

Called at the end of fit (train + validate), validate, test, or predict.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

test_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for testing.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• test()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying testing samples.

Example:

def test_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def test_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a test dataset and a test_step(), you don’t need to implement this method.

Note

In the case where you return multiple test dataloaders, the test_step() will have an argument dataloader_idx which matches the order here.

test_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of a test epoch with the output of all test steps.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

outputs – List of outputs you defined in test_step_end(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader

Returns

None

Note

If you didn’t define a test_step(), this won’t be called.

Examples

With a single dataloader:

def test_epoch_end(self, outputs):
    # do something with the outputs of all test batches
    all_test_preds = test_step_outputs.predictions

    some_result = calc_all_results(all_test_preds)
    self.log(some_result)

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each test step for that dataloader.

def test_epoch_end(self, outputs):
    final_value = 0
    for dataloader_outputs in outputs:
        for test_step_out in dataloader_outputs:
            # do something
            final_value += test_step_out

    self.log("final_metric", final_value)

test_step(batch, batch_idx)

Operates on a single batch of data from the test set. In this step you’d normally generate examples or calculate anything of interest such as accuracy.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_id – The index of the dataloader that produced this batch. (only if multiple test dataloaders used).

Returns

Any of.

• Any object or value

• None - Testing will skip to the next batch

# if you have one test dataloader:
def test_step(self, batch, batch_idx):
    ...


# if you have multiple test dataloaders:
def test_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single test dataset
def test_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    test_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'test_loss': loss, 'test_acc': test_acc})

If you pass in multiple test dataloaders, test_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple test dataloaders
def test_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to test you don’t need to implement this method.

Note

When the test_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of the test epoch, the model goes back to training mode and gradients are enabled.

test_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when testing with DP because test_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [test_step(sub_batch) for sub_batch in sub_batches]
test_step_end(step_output)

Parameters

step_output – What you return in test_step() for each batch part.

Returns

None or anything

# WITHOUT test_step_end
# if used in DP, this batch is 1/num_gpus large
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    loss = self.softmax(out)
    self.log("test_loss", loss)


# --------------
# with test_step_end to do softmax over the full batch
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return out


def test_step_end(self, output_results):
    # this out is now the full size of the batch
    all_test_step_outs = output_results.out
    loss = nce_loss(all_test_step_outs)
    self.log("test_loss", loss)

See also

See the Multi GPU Training guide for more details.

to(*args: Any, **kwargs: Any) → Self

See torch.nn.Module.to().

to_empty(*, device: Union[str, device]) → T

Moves the parameters and buffers to the specified device without copying storage.

Parameters

device (torch.device) – The desired device of the parameters and buffers in this module.

Returns

self

Return type

Module

to_onnx(file_path: Union[str, Path], input_sample: Optional[Any] = None, **kwargs: Any) → None

Saves the model in ONNX format.

Parameters

• file_path – The path of the file the onnx model should be saved to.

• input_sample – An input for tracing. Default: None (Use self.example_input_array)

• **kwargs – Will be passed to torch.onnx.export function.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))

>>> with tempfile.NamedTemporaryFile(suffix='.onnx', delete=False) as tmpfile:
...     model = SimpleModel()
...     input_sample = torch.randn((1, 64))
...     model.to_onnx(tmpfile.name, input_sample, export_params=True)
...     os.path.isfile(tmpfile.name)
True

to_torchscript(file_path: Optional[Union[str, Path]] = None, method: Optional[str] = 'script', example_inputs: Optional[Any] = None, **kwargs: Any) → Union[ScriptModule, Dict[str, ScriptModule]]

By default compiles the whole model to a ScriptModule. If you want to use tracing, please provided the argument method='trace' and make sure that either the example_inputs argument is provided, or the model has example_input_array set. If you would like to customize the modules that are scripted you should override this method. In case you want to return multiple modules, we recommend using a dictionary.

Parameters

• file_path – Path where to save the torchscript. Default: None (no file saved).

• method – Whether to use TorchScript’s script or trace method. Default: ‘script’

• example_inputs – An input to be used to do tracing when method is set to ‘trace’. Default: None (uses example_input_array)

• **kwargs – Additional arguments that will be passed to the torch.jit.script() or torch.jit.trace() function.

Note

• Requires the implementation of the forward() method.

• The exported script will be set to evaluation mode.

• It is recommended that you install the latest supported version of PyTorch to use this feature without limitations. See also the torch.jit documentation for supported features.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))
...
>>> model = SimpleModel()
>>> model.to_torchscript(file_path="model.pt")  
>>> os.path.isfile("model.pt")  
>>> torch.jit.save(model.to_torchscript(file_path="model_trace.pt", method='trace', 
...                                     example_inputs=torch.randn(1, 64)))  
>>> os.path.isfile("model_trace.pt")  
True

Returns

This LightningModule as a torchscript, regardless of whether file_path is defined or not.

toggle_optimizer(optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int) → None

Makes sure only the gradients of the current optimizer’s parameters are calculated in the training step to prevent dangling gradients in multiple-optimizer setup.

This is only called automatically when automatic optimization is enabled and multiple optimizers are used. It works with untoggle_optimizer() to make sure param_requires_grad_state is properly reset.

Parameters

• optimizer – The optimizer to toggle.

• optimizer_idx – The index of the optimizer to toggle.

train(mode: bool = True) → T

Sets the module in training mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

Parameters

mode (bool) – whether to set training mode (True) or evaluation mode (False). Default: True.

Returns

self

Return type

Module

train_dataloader() → Union[DataLoader, Sequence[DataLoader], Sequence[Sequence[DataLoader]], Sequence[Dict[str, DataLoader]], Dict[str, DataLoader], Dict[str, Dict[str, DataLoader]], Dict[str, Sequence[DataLoader]]]

Implement one or more PyTorch DataLoaders for training.

Returns

A collection of torch.utils.data.DataLoader specifying training samples. In the case of multiple dataloaders, please see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• fit()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Example:

# single dataloader
def train_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=True, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=True
    )
    return loader

# multiple dataloaders, return as list
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a list of tensors: [batch_mnist, batch_cifar]
    return [mnist_loader, cifar_loader]

# multiple dataloader, return as dict
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a dict of tensors: {'mnist': batch_mnist, 'cifar': batch_cifar}
    return {'mnist': mnist_loader, 'cifar': cifar_loader}

training_epoch_end(outputs: List[Union[Tensor, Dict[str, Any]]]) → None

Called at the end of the training epoch with the outputs of all training steps. Use this in case you need to do something with all the outputs returned by training_step().

# the pseudocode for these calls
train_outs = []
for train_batch in train_data:
    out = training_step(train_batch)
    train_outs.append(out)
training_epoch_end(train_outs)

Parameters

outputs – List of outputs you defined in training_step(). If there are multiple optimizers or when using truncated_bptt_steps > 0, the lists have the dimensions (n_batches, tbptt_steps, n_optimizers). Dimensions of length 1 are squeezed.

Returns

None

Note

If this method is not overridden, this won’t be called.

def training_epoch_end(self, training_step_outputs):
    # do something with all training_step outputs
    for out in training_step_outputs:
        ...

training_step(batch, batch_idx)

Here you compute and return the training loss and some additional metrics for e.g. the progress bar or logger.

Parameters

• batch (Tensor | (Tensor, …) | [Tensor, …]) – The output of your DataLoader. A tensor, tuple or list.

• batch_idx (int) – Integer displaying index of this batch

• optimizer_idx (int) – When using multiple optimizers, this argument will also be present.

• hiddens (Any) – Passed in if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0.

Returns

Any of.

• Tensor - The loss tensor

• dict - A dictionary. Can include any keys, but must include the key 'loss'

• None - Training will skip to the next batch. This is only for automatic optimization.

  This is not supported for multi-GPU, TPU, IPU, or DeepSpeed.

In this step you’d normally do the forward pass and calculate the loss for a batch. You can also do fancier things like multiple forward passes or something model specific.

Example:

def training_step(self, batch, batch_idx):
    x, y, z = batch
    out = self.encoder(x)
    loss = self.loss(out, x)
    return loss

If you define multiple optimizers, this step will be called with an additional optimizer_idx parameter.

# Multiple optimizers (e.g.: GANs)
def training_step(self, batch, batch_idx, optimizer_idx):
    if optimizer_idx == 0:
        # do training_step with encoder
        ...
    if optimizer_idx == 1:
        # do training_step with decoder
        ...

If you add truncated back propagation through time you will also get an additional argument with the hidden states of the previous step.

# Truncated back-propagation through time
def training_step(self, batch, batch_idx, hiddens):
    # hiddens are the hidden states from the previous truncated backprop step
    out, hiddens = self.lstm(data, hiddens)
    loss = ...
    return {"loss": loss, "hiddens": hiddens}

Note

The loss value shown in the progress bar is smoothed (averaged) over the last values, so it differs from the actual loss returned in train/validation step.

Note

When accumulate_grad_batches > 1, the loss returned here will be automatically normalized by accumulate_grad_batches internally.

training_step_end(step_output: Union[Tensor, Dict[str, Any]]) → Union[Tensor, Dict[str, Any]]

Use this when training with dp because training_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [training_step(sub_batch) for sub_batch in sub_batches]
training_step_end(step_output)

Parameters

step_output – What you return in training_step for each batch part.

Returns

Anything

When using the DP strategy, only a portion of the batch is inside the training_step:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)

    # softmax uses only a portion of the batch in the denominator
    loss = self.softmax(out)
    loss = nce_loss(loss)
    return loss

If you wish to do something with all the parts of the batch, then use this method to do it:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return {"pred": out}


def training_step_end(self, training_step_outputs):
    gpu_0_pred = training_step_outputs[0]["pred"]
    gpu_1_pred = training_step_outputs[1]["pred"]
    gpu_n_pred = training_step_outputs[n]["pred"]

    # this softmax now uses the full batch
    loss = nce_loss([gpu_0_pred, gpu_1_pred, gpu_n_pred])
    return loss

See also

See the Multi GPU Training guide for more details.

transfer_batch_to_device(batch: Any, device: device, dataloader_idx: int) → Any

Override this hook if your DataLoader returns tensors wrapped in a custom data structure.

The data types listed below (and any arbitrary nesting of them) are supported out of the box:

• torch.Tensor or anything that implements .to(…)

• list

• dict

• tuple

For anything else, you need to define how the data is moved to the target device (CPU, GPU, TPU, …).

Note

This hook should only transfer the data and not modify it, nor should it move the data to any other device than the one passed in as argument (unless you know what you are doing). To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be transferred to a new device.

• device – The target device as defined in PyTorch.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A reference to the data on the new device.

Example:

def transfer_batch_to_device(self, batch, device, dataloader_idx):
    if isinstance(batch, CustomBatch):
        # move all tensors in your custom data structure to the device
        batch.samples = batch.samples.to(device)
        batch.targets = batch.targets.to(device)
    elif dataloader_idx == 0:
        # skip device transfer for the first dataloader or anything you wish
        pass
    else:
        batch = super().transfer_batch_to_device(data, device, dataloader_idx)
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• move_data_to_device()

• apply_to_collection()

transform(loader: DataLoader)

Parameters

loader – a dataloader that matches the structure of that used for training

Returns

transformed views

property truncated_bptt_steps: int

Enables Truncated Backpropagation Through Time in the Trainer when set to a positive integer.

It represents the number of times training_step() gets called before backpropagation. If this is > 0, the training_step() receives an additional argument hiddens and is expected to return a hidden state.

type(dst_type: Union[str, dtype]) → Self

See torch.nn.Module.type().

unfreeze() → None

Unfreeze all parameters for training.

model = MyLightningModule(...)
model.unfreeze()

untoggle_optimizer(optimizer_idx: int) → None

Resets the state of required gradients that were toggled with toggle_optimizer().

This is only called automatically when automatic optimization is enabled and multiple optimizers are used.

Parameters

optimizer_idx – The index of the optimizer to untoggle.

val_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for validation.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

It’s recommended that all data downloads and preparation happen in prepare_data().

• fit()

• validate()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying validation samples.

Examples:

def val_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False,
                    transform=transform, download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def val_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a validation dataset and a validation_step(), you don’t need to implement this method.

Note

In the case where you return multiple validation dataloaders, the validation_step() will have an argument dataloader_idx which matches the order here.

validation_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of the validation epoch with the outputs of all validation steps.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

outputs – List of outputs you defined in validation_step(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader.

Returns

None

Note

If you didn’t define a validation_step(), this won’t be called.

Examples

With a single dataloader:

def validation_epoch_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each validation step for that dataloader.

def validation_epoch_end(self, outputs):
    for dataloader_output_result in outputs:
        dataloader_outs = dataloader_output_result.dataloader_i_outputs

    self.log("final_metric", final_value)

validation_step(batch, batch_idx)

Operates on a single batch of data from the validation set. In this step you’d might generate examples or calculate anything of interest like accuracy.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_idx – The index of the dataloader that produced this batch. (only if multiple val dataloaders used)

Returns

• Any object or value

• None - Validation will skip to the next batch

# pseudocode of order
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    if defined("validation_step_end"):
        out = validation_step_end(out)
    val_outs.append(out)
val_outs = validation_epoch_end(val_outs)

# if you have one val dataloader:
def validation_step(self, batch, batch_idx):
    ...


# if you have multiple val dataloaders:
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single validation dataset
def validation_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    val_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'val_loss': loss, 'val_acc': val_acc})

If you pass in multiple val dataloaders, validation_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple validation dataloaders
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to validate you don’t need to implement this method.

Note

When the validation_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of validation, the model goes back to training mode and gradients are enabled.

validation_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when validating with dp because validation_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [validation_step(sub_batch) for sub_batch in sub_batches]
validation_step_end(step_output)

Parameters

step_output – What you return in validation_step() for each batch part.

Returns

None or anything

# WITHOUT validation_step_end
# if used in DP, this batch is 1/num_gpus large
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    loss = self.softmax(out)
    loss = nce_loss(loss)
    self.log("val_loss", loss)


# --------------
# with validation_step_end to do softmax over the full batch
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    return out


def validation_step_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

See also

See the Multi GPU Training guide for more details.

xpu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the XPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

zero_grad(set_to_none: bool = False) → None

Sets gradients of all model parameters to zero. See similar function under torch.optim.Optimizer for more context.

Parameters

set_to_none (bool) – instead of setting to zero, set the grads to None. See torch.optim.Optimizer.zero_grad() for details.

class cca_zoo.deepmodels.DVCCA(latent_dims: int, encoders=None, decoders=None, private_encoders: Optional[Iterable] = None, latent_dropout=0, img_dim=None, recon_loss_type='mse', **kwargs)

Bases: _BaseDeep, _GenerativeMixin

A class used to fit a DVCCA model.

References

Wang, Weiran, et al. ‘Deep variational canonical correlation analysis.’ arXiv preprint arXiv:1610.03454 (2016). https: // arxiv.org / pdf / 1610.03454.pdf https: // github.com / pytorch / examples / blob / master / vae / main.py

forward(views, mle=True, **kwargs)

Forward method for the model. Outputs latent encoding for each view

Parameters

• views –

• kwargs –

Returns

loss(views, **kwargs)

Loss for model :param views: :param args: :param kwargs: :return:

transform(loader: DataLoader)

Parameters

loader – a dataloader that matches the structure of that used for training

Returns

transformed views

add_module(name: str, module: Optional[Module]) → None

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Parameters

• name (str) – name of the child module. The child module can be accessed from this module using the given name

• module (Module) – child module to be added to the module.

all_gather(data: Union[Tensor, Dict, List, Tuple], group: Optional[Any] = None, sync_grads: bool = False) → Union[Tensor, Dict, List, Tuple]

Allows users to call self.all_gather() from the LightningModule, thus making the all_gather operation accelerator agnostic. all_gather is a function provided by accelerators to gather a tensor from several distributed processes.

Parameters

• data – int, float, tensor of shape (batch, …), or a (possibly nested) collection thereof.

• group – the process group to gather results from. Defaults to all processes (world)

• sync_grads – flag that allows users to synchronize gradients for the all_gather operation

Returns

A tensor of shape (world_size, batch, …), or if the input was a collection the output will also be a collection with tensors of this shape.

apply(fn: Callable[[Module], None]) → T

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also torch.nn.init).

Parameters

fn (Module -> None) – function to be applied to each submodule

Returns

self

Return type

Module

Example:

>>> @torch.no_grad()
>>> def init_weights(m):
>>>     print(m)
>>>     if type(m) == nn.Linear:
>>>         m.weight.fill_(1.0)
>>>         print(m.weight)
>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

property automatic_optimization: bool

If set to False you are responsible for calling .backward(), .step(), .zero_grad().

backward(loss: Tensor, optimizer: Optional[Steppable], optimizer_idx: Optional[int], *args: Any, **kwargs: Any) → None

Called to perform backward on the loss returned in training_step(). Override this hook with your own implementation if you need to.

Parameters

• loss – The loss tensor returned by training_step(). If gradient accumulation is used, the loss here holds the normalized value (scaled by 1 / accumulation steps).

• optimizer – Current optimizer being used. None if using manual optimization.

• optimizer_idx – Index of the current optimizer being used. None if using manual optimization.

Example:

def backward(self, loss, optimizer, optimizer_idx):
    loss.backward()

bfloat16() → T

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns

self

Return type

Module

buffers(recurse: bool = True) → Iterator[Tensor]

Returns an iterator over module buffers.

Parameters

recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

torch.Tensor – module buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for buf in model.buffers():
>>>     print(type(buf), buf.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

children() → Iterator[Module]

Returns an iterator over immediate children modules.

Yields

Module – a child module

clip_gradients(optimizer: Optimizer, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Handles gradient clipping internally.

Note

Do not override this method. If you want to customize gradient clipping, consider using configure_gradient_clipping() method.

Parameters

• optimizer – Current optimizer being used.

• gradient_clip_val – The value at which to clip gradients.

• gradient_clip_algorithm – The gradient clipping algorithm to use. Pass gradient_clip_algorithm="value" to clip by value, and gradient_clip_algorithm="norm" to clip by norm.

configure_callbacks()

Configure model-specific callbacks. When the model gets attached, e.g., when .fit() or .test() gets called, the list or a callback returned here will be merged with the list of callbacks passed to the Trainer’s callbacks argument. If a callback returned here has the same type as one or several callbacks already present in the Trainer’s callbacks list, it will take priority and replace them. In addition, Lightning will make sure ModelCheckpoint callbacks run last.

Returns

A callback or a list of callbacks which will extend the list of callbacks in the Trainer.

Example:

def configure_callbacks(self):
    early_stop = EarlyStopping(monitor="val_acc", mode="max")
    checkpoint = ModelCheckpoint(monitor="val_loss")
    return [early_stop, checkpoint]

configure_gradient_clipping(optimizer: Optimizer, optimizer_idx: int, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Perform gradient clipping for the optimizer parameters. Called before optimizer_step().

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

• gradient_clip_val – The value at which to clip gradients. By default value passed in Trainer will be available here.

• gradient_clip_algorithm – The gradient clipping algorithm to use. By default value passed in Trainer will be available here.

Example:

# Perform gradient clipping on gradients associated with discriminator (optimizer_idx=1) in GAN
def configure_gradient_clipping(self, optimizer, optimizer_idx, gradient_clip_val, gradient_clip_algorithm):
    if optimizer_idx == 1:
        # Lightning will handle the gradient clipping
        self.clip_gradients(
            optimizer,
            gradient_clip_val=gradient_clip_val,
            gradient_clip_algorithm=gradient_clip_algorithm
        )
    else:
        # implement your own custom logic to clip gradients for generator (optimizer_idx=0)

configure_optimizers()

Collects learnable parameters and configures the optimizer and learning rate scheduler. :returns: two lists containing the optimizer and the scheduler. :rtype: Tuple[List, List]

configure_sharded_model() → None

Hook to create modules in a distributed aware context. This is useful for when using sharded plugins, where we’d like to shard the model instantly, which is useful for extremely large models which can save memory and initialization time.

This hook is called during each of fit/val/test/predict stages in the same process, so ensure that implementation of this hook is idempotent.

cpu() → Self

See torch.nn.Module.cpu().

cuda(device: Optional[Union[device, int]] = None) → Self

Moves all model parameters and buffers to the GPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized.

Parameters

device – If specified, all parameters will be copied to that device. If None, the current CUDA device index will be used.

Returns

self

Return type

Module

property current_epoch: int

The current epoch in the Trainer, or 0 if not attached.

double() → Self

See torch.nn.Module.double().

eval() → T

Sets the module in evaluation mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False).

See Locally disabling gradient computation for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns

self

Return type

Module

property example_input_array: Optional[Union[Tensor, Tuple, Dict]]

The example input array is a specification of what the module can consume in the forward() method. The return type is interpreted as follows:

• Single tensor: It is assumed the model takes a single argument, i.e., model.forward(model.example_input_array)

• Tuple: The input array should be interpreted as a sequence of positional arguments, i.e., model.forward(*model.example_input_array)

• Dict: The input array represents named keyword arguments, i.e., model.forward(**model.example_input_array)

extra_repr() → str

Set the extra representation of the module

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float() → Self

See torch.nn.Module.float().

freeze() → None

Freeze all params for inference.

Example:

model = MyLightningModule(...)
model.freeze()

get_buffer(target: str) → Tensor

Returns the buffer given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The buffer referenced by target

Return type

torch.Tensor

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state() → Any

Returns any extra state to include in the module’s state_dict. Implement this and a corresponding set_extra_state() for your module if you need to store extra state. This function is called when building the module’s state_dict().

Note that extra state should be pickleable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns

Any extra state to store in the module’s state_dict

Return type

object

get_parameter(target: str) → Parameter

Returns the parameter given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The Parameter referenced by target

Return type

torch.nn.Parameter

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(target: str) → Module

Returns the submodule given by target if it exists, otherwise throws an error.

For example, let’s say you have an nn.Module A that looks like this:

A(
    (net_b): Module(
        (net_c): Module(
            (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2))
        )
        (linear): Linear(in_features=100, out_features=200, bias=True)
    )
)

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Parameters

target – The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns

The submodule referenced by target

Return type

torch.nn.Module

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Module

property global_rank: int

The index of the current process across all nodes and devices.

property global_step: int

Total training batches seen across all epochs.

If no Trainer is attached, this propery is 0.

half() → Self

See torch.nn.Module.half().

property hparams: Union[AttributeDict, MutableMapping]

The collection of hyperparameters saved with save_hyperparameters(). It is mutable by the user. For the frozen set of initial hyperparameters, use hparams_initial.

Returns

Mutable hyperparameters dictionary

property hparams_initial: AttributeDict

The collection of hyperparameters saved with save_hyperparameters(). These contents are read-only. Manual updates to the saved hyperparameters can instead be performed through hparams.

Returns

immutable initial hyperparameters

Return type

AttributeDict

ipu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the IPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on IPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

classmethod load_from_checkpoint(checkpoint_path: Union[str, Path, IO], map_location: Optional[Union[device, str, int, Callable[[Union[device, str, int]], Union[device, str, int]], Dict[Union[device, str, int], Union[device, str, int]]]] = None, hparams_file: Optional[Union[str, Path]] = None, strict: bool = True, **kwargs: Any) → Self

Primary way of loading a model from a checkpoint. When Lightning saves a checkpoint it stores the arguments passed to __init__ in the checkpoint under "hyper_parameters".

Any arguments specified through **kwargs will override args stored in "hyper_parameters".

Parameters

• checkpoint_path – Path to checkpoint. This can also be a URL, or file-like object

• map_location – If your checkpoint saved a GPU model and you now load on CPUs or a different number of GPUs, use this to map to the new setup. The behaviour is the same as in torch.load().

• hparams_file –

  Optional path to a .yaml or .csv file with hierarchical structure as in this example:

  drop_prob: 0.2
  dataloader:
      batch_size: 32
  

  You most likely won’t need this since Lightning will always save the hyperparameters to the checkpoint. However, if your checkpoint weights don’t have the hyperparameters saved, use this method to pass in a .yaml file with the hparams you’d like to use. These will be converted into a dict and passed into your LightningModule for use.

  If your model’s hparams argument is Namespace and .yaml file has hierarchical structure, you need to refactor your model to treat hparams as dict.

• strict – Whether to strictly enforce that the keys in checkpoint_path match the keys returned by this module’s state dict.

• **kwargs – Any extra keyword args needed to init the model. Can also be used to override saved hyperparameter values.

Returns

LightningModule instance with loaded weights and hyperparameters (if available).

Note

load_from_checkpoint is a class method. You should use your LightningModule class to call it instead of the LightningModule instance.

Example:

# load weights without mapping ...
model = MyLightningModule.load_from_checkpoint('path/to/checkpoint.ckpt')

# or load weights mapping all weights from GPU 1 to GPU 0 ...
map_location = {'cuda:1':'cuda:0'}
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    map_location=map_location
)

# or load weights and hyperparameters from separate files.
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    hparams_file='/path/to/hparams_file.yaml'
)

# override some of the params with new values
model = MyLightningModule.load_from_checkpoint(
    PATH,
    num_layers=128,
    pretrained_ckpt_path=NEW_PATH,
)

# predict
pretrained_model.eval()
pretrained_model.freeze()
y_hat = pretrained_model(x)

load_state_dict(state_dict: Mapping[str, Any], strict: bool = True)

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Parameters

• state_dict (dict) – a dict containing parameters and persistent buffers.

• strict (bool, optional) – whether to strictly enforce that the keys in state_dict match the keys returned by this module’s state_dict() function. Default: True

Returns

• missing_keys is a list of str containing the missing keys

• unexpected_keys is a list of str containing the unexpected keys

Return type

NamedTuple with missing_keys and unexpected_keys fields

Note

If a parameter or buffer is registered as None and its corresponding key exists in state_dict, load_state_dict() will raise a RuntimeError.

property local_rank: int

The index of the current process within a single node.

log(name: str, value: Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, metric_attribute: Optional[str] = None, rank_zero_only: bool = False) → None

Log a key, value pair.

Example:

self.log('train_loss', loss)

The default behavior per hook is documented here: extensions/logging:Automatic Logging.

Parameters

• name – key to log.

• value – value to log. Can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress bar.

• logger – if True logs to the logger.

• on_step – if True logs at this step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto detach the graph.

• sync_dist – if True, reduces the metric across devices. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the DDP group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple dataloaders). If False, user needs to give unique names for each dataloader to not mix the values.

• batch_size – Current batch_size. This will be directly inferred from the loaded batch, but for some data structures you might need to explicitly provide it.

• metric_attribute – To restore the metric state, Lightning requires the reference of the torchmetrics.Metric in your model. This is found automatically if it is a model attribute.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_dict(dictionary: Mapping[str, Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, rank_zero_only: bool = False) → None

Log a dictionary of values at once.

Example:

values = {'loss': loss, 'acc': acc, ..., 'metric_n': metric_n}
self.log_dict(values)

Parameters

• dictionary – key value pairs. The values can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress base.

• logger – if True logs to the logger.

• on_step – if True logs at this step. None auto-logs for training_step but not validation/test_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. None auto-logs for val/test step but not training_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto-detach the graph

• sync_dist – if True, reduces the metric across GPUs/TPUs. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the ddp group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple). If False, user needs to give unique names for each dataloader to not mix values.

• batch_size – Current batch size. This will be directly inferred from the loaded batch, but some data structures might need to explicitly provide it.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_grad_norm(grad_norm_dict: Dict[str, float]) → None

Override this method to change the default behaviour of log_grad_norm.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

grad_norm_dict – Dictionary containing current grad norm metrics

Example:

# DEFAULT
def log_grad_norm(self, grad_norm_dict):
    self.log_dict(grad_norm_dict, on_step=True, on_epoch=True, prog_bar=False, logger=True)

property logger: Optional[Logger]

Reference to the logger object in the Trainer.

property loggers: List[Logger]

Reference to the list of loggers in the Trainer.

lr_scheduler_step(scheduler: Union[_LRScheduler, ReduceLROnPlateau], optimizer_idx: int, metric: Optional[Any]) → None

Override this method to adjust the default way the Trainer calls each scheduler. By default, Lightning calls step() and as shown in the example for each scheduler based on its interval.

Parameters

• scheduler – Learning rate scheduler.

• optimizer_idx – Index of the optimizer associated with this scheduler.

• metric – Value of the monitor used for schedulers like ReduceLROnPlateau.

Examples:

# DEFAULT
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    if metric is None:
        scheduler.step()
    else:
        scheduler.step(metric)

# Alternative way to update schedulers if it requires an epoch value
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    scheduler.step(epoch=self.current_epoch)

lr_schedulers() → Union[None, List[Union[_LRScheduler, ReduceLROnPlateau]], _LRScheduler, ReduceLROnPlateau]

Returns the learning rate scheduler(s) that are being used during training. Useful for manual optimization.

Returns

A single scheduler, or a list of schedulers in case multiple ones are present, or None if no schedulers were returned in configure_optimizers().

manual_backward(loss: Tensor, *args: Any, **kwargs: Any) → None

Call this directly from your training_step() when doing optimizations manually. By using this, Lightning can ensure that all the proper scaling gets applied when using mixed precision.

See manual optimization for more examples.

Example:

def training_step(...):
    opt = self.optimizers()
    loss = ...
    opt.zero_grad()
    # automatically applies scaling, etc...
    self.manual_backward(loss)
    opt.step()

Parameters

• loss – The tensor on which to compute gradients. Must have a graph attached.

• *args – Additional positional arguments to be forwarded to backward()

• **kwargs – Additional keyword arguments to be forwarded to backward()

modules() → Iterator[Module]

Returns an iterator over all modules in the network.

Yields

Module – a module in the network

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.modules()):
...     print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Tensor]]

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Parameters

• prefix (str) – prefix to prepend to all buffer names.

• recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

(str, torch.Tensor) – Tuple containing the name and buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, buf in self.named_buffers():
>>>    if name in ['running_var']:
>>>        print(buf.size())

named_children() → Iterator[Tuple[str, Module]]

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields

(str, Module) – Tuple containing a name and child module

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, module in model.named_children():
>>>     if name in ['conv4', 'conv5']:
>>>         print(module)

named_modules(memo: Optional[Set[Module]] = None, prefix: str = '', remove_duplicate: bool = True)

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Parameters

• memo – a memo to store the set of modules already added to the result

• prefix – a prefix that will be added to the name of the module

• remove_duplicate – whether to remove the duplicated module instances in the result or not

Yields

(str, Module) – Tuple of name and module

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.named_modules()):
...     print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Parameter]]

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Parameters

• prefix (str) – prefix to prepend to all parameter names.

• recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

(str, Parameter) – Tuple containing the name and parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, param in self.named_parameters():
>>>    if name in ['bias']:
>>>        print(param.size())

on_after_backward() → None

Called after loss.backward() and before optimizers are stepped.

Note

If using native AMP, the gradients will not be unscaled at this point. Use the on_before_optimizer_step if you need the unscaled gradients.

on_after_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch after it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_after_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = gpu_transforms(batch['x'])
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• on_before_batch_transfer()

• transfer_batch_to_device()

on_before_backward(loss: Tensor) → None

Called before loss.backward().

Parameters

loss – Loss divided by number of batches for gradient accumulation and scaled if using native AMP.

on_before_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch before it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_before_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = transforms(batch['x'])
    return batch

See also

• on_after_batch_transfer()

• transfer_batch_to_device()

on_before_optimizer_step(optimizer: Optimizer, optimizer_idx: int) → None

Called before optimizer.step().

If using gradient accumulation, the hook is called once the gradients have been accumulated. See: :paramref:`~pytorch_lightning.trainer.Trainer.accumulate_grad_batches`.

If using native AMP, the loss will be unscaled before calling this hook. See these docs for more information on the scaling of gradients.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

Example:

def on_before_optimizer_step(self, optimizer, optimizer_idx):
    # example to inspect gradient information in tensorboard
    if self.trainer.global_step % 25 == 0:  # don't make the tf file huge
        for k, v in self.named_parameters():
            self.logger.experiment.add_histogram(
                tag=k, values=v.grad, global_step=self.trainer.global_step
            )

on_before_zero_grad(optimizer: Optimizer) → None

Called after training_step() and before optimizer.zero_grad().

Called in the training loop after taking an optimizer step and before zeroing grads. Good place to inspect weight information with weights updated.

This is where it is called:

for optimizer in optimizers:
    out = training_step(...)

    model.on_before_zero_grad(optimizer) # < ---- called here
    optimizer.zero_grad()

    backward()

Parameters

optimizer – The optimizer for which grads should be zeroed.

on_fit_end() → None

Called at the very end of fit.

If on DDP it is called on every process

on_fit_start() → None

Called at the very beginning of fit.

If on DDP it is called on every process

property on_gpu: bool

Returns True if this model is currently located on a GPU.

Useful to set flags around the LightningModule for different CPU vs GPU behavior.

on_load_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning to restore your model. If you saved something with on_save_checkpoint() this is your chance to restore this.

Parameters

checkpoint – Loaded checkpoint

Example:

def on_load_checkpoint(self, checkpoint):
    # 99% of the time you don't need to implement this method
    self.something_cool_i_want_to_save = checkpoint['something_cool_i_want_to_save']

Note

Lightning auto-restores global step, epoch, and train state including amp scaling. There is no need for you to restore anything regarding training.

on_predict_batch_end(outputs: Optional[Any], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop after the batch.

Parameters

• outputs – The outputs of predict_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_end() → None

Called at the end of predicting.

on_predict_epoch_end(results: List[Any]) → None

Called at the end of predicting.

on_predict_epoch_start() → None

Called at the beginning of predicting.

on_predict_model_eval() → None

Sets the model to eval during the predict loop.

on_predict_start() → None

Called at the beginning of predicting.

on_save_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning when saving a checkpoint to give you a chance to store anything else you might want to save.

Parameters

checkpoint – The full checkpoint dictionary before it gets dumped to a file. Implementations of this hook can insert additional data into this dictionary.

Example:

def on_save_checkpoint(self, checkpoint):
    # 99% of use cases you don't need to implement this method
    checkpoint['something_cool_i_want_to_save'] = my_cool_pickable_object

Note

Lightning saves all aspects of training (epoch, global step, etc…) including amp scaling. There is no need for you to store anything about training.

on_test_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop after the batch.

Parameters

• outputs – The outputs of test_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_end() → None

Called at the end of testing.

on_test_epoch_end() → None

Called in the test loop at the very end of the epoch.

on_test_epoch_start() → None

Called in the test loop at the very beginning of the epoch.

on_test_model_eval() → None

Sets the model to eval during the test loop.

on_test_model_train() → None

Sets the model to train during the test loop.

on_test_start() → None

Called at the beginning of testing.

on_train_batch_end(outputs: Union[Tensor, Dict[str, Any]], batch: Any, batch_idx: int) → None

Called in the training loop after the batch.

Parameters

• outputs – The outputs of training_step_end(training_step(x))

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_batch_start(batch: Any, batch_idx: int) → Optional[int]

Called in the training loop before anything happens for that batch.

If you return -1 here, you will skip training for the rest of the current epoch.

Parameters

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_end() → None

Called at the end of training before logger experiment is closed.

on_train_epoch_end() → None

Called in the training loop at the very end of the epoch.

To access all batch outputs at the end of the epoch, either:

1. Implement training_epoch_end in the LightningModule OR

2. Cache data across steps on the attribute(s) of the LightningModule and access them in this hook

on_train_epoch_start() → None

Called in the training loop at the very beginning of the epoch.

on_train_start() → None

Called at the beginning of training after sanity check.

on_validation_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop after the batch.

Parameters

• outputs – The outputs of validation_step_end(validation_step(x))

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_end() → None

Called at the end of validation.

on_validation_epoch_end() → None

Called in the validation loop at the very end of the epoch.

on_validation_epoch_start() → None

Called in the validation loop at the very beginning of the epoch.

on_validation_model_eval() → None

Sets the model to eval during the val loop.

on_validation_model_train() → None

Sets the model to train during the val loop.

on_validation_start() → None

Called at the beginning of validation.

optimizer_step(epoch: int, batch_idx: int, optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int = 0, optimizer_closure: Optional[Callable[[], Any]] = None, on_tpu: bool = False, using_native_amp: bool = False, using_lbfgs: bool = False) → None

Override this method to adjust the default way the Trainer calls each optimizer.

By default, Lightning calls step() and zero_grad() as shown in the example once per optimizer. This method (and zero_grad()) won’t be called during the accumulation phase when Trainer(accumulate_grad_batches != 1). Overriding this hook has no benefit with manual optimization.

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers, this indexes into that list.

• optimizer_closure – The optimizer closure. This closure must be executed as it includes the calls to training_step(), optimizer.zero_grad(), and backward().

• on_tpu – True if TPU backward is required

• using_native_amp – True if using native amp

• using_lbfgs – True if the matching optimizer is torch.optim.LBFGS

Examples:

# DEFAULT
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    optimizer.step(closure=optimizer_closure)

# Alternating schedule for optimizer steps (i.e.: GANs)
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    # update generator opt every step
    if optimizer_idx == 0:
        optimizer.step(closure=optimizer_closure)

    # update discriminator opt every 2 steps
    if optimizer_idx == 1:
        if (batch_idx + 1) % 2 == 0 :
            optimizer.step(closure=optimizer_closure)
        else:
            # call the closure by itself to run `training_step` + `backward` without an optimizer step
            optimizer_closure()

    # ...
    # add as many optimizers as you want

Here’s another example showing how to use this for more advanced things such as learning rate warm-up:

# learning rate warm-up
def optimizer_step(
    self,
    epoch,
    batch_idx,
    optimizer,
    optimizer_idx,
    optimizer_closure,
    on_tpu,
    using_native_amp,
    using_lbfgs,
):
    # update params
    optimizer.step(closure=optimizer_closure)

    # manually warm up lr without a scheduler
    if self.trainer.global_step < 500:
        lr_scale = min(1.0, float(self.trainer.global_step + 1) / 500.0)
        for pg in optimizer.param_groups:
            pg["lr"] = lr_scale * self.learning_rate

optimizer_zero_grad(epoch: int, batch_idx: int, optimizer: Optimizer, optimizer_idx: int) → None

Override this method to change the default behaviour of optimizer.zero_grad().

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers this indexes into that list.

Examples:

# DEFAULT
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad()

# Set gradients to `None` instead of zero to improve performance.
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad(set_to_none=True)

See torch.optim.Optimizer.zero_grad() for the explanation of the above example.

optimizers(use_pl_optimizer: bool = True) → Union[Optimizer, LightningOptimizer, List[Optimizer], List[LightningOptimizer]]

Returns the optimizer(s) that are being used during training. Useful for manual optimization.

Parameters

use_pl_optimizer – If True, will wrap the optimizer(s) in a LightningOptimizer for automatic handling of precision and profiling.

Returns

A single optimizer, or a list of optimizers in case multiple ones are present.

parameters(recurse: bool = True) → Iterator[Parameter]

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Parameters

recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

Parameter – module parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for param in model.parameters():
>>>     print(type(param), param.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

predict_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for prediction.

It’s recommended that all data downloads and preparation happen in prepare_data().

• predict()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying prediction samples.

Note

In the case where you return multiple prediction dataloaders, the predict_step() will have an argument dataloader_idx which matches the order here.

predict_step(batch: Any, batch_idx: int, dataloader_idx: int = 0) → Any

Step function called during predict(). By default, it calls forward(). Override to add any processing logic.

The predict_step() is used to scale inference on multi-devices.

To prevent an OOM error, it is possible to use BasePredictionWriter callback to write the predictions to disk or database after each batch or on epoch end.

The BasePredictionWriter should be used while using a spawn based accelerator. This happens for Trainer(strategy="ddp_spawn") or training on 8 TPU cores with Trainer(accelerator="tpu", devices=8) as predictions won’t be returned.

Example

class MyModel(LightningModule):

    def predict_step(self, batch, batch_idx, dataloader_idx=0):
        return self(batch)

dm = ...
model = MyModel()
trainer = Trainer(accelerator="gpu", devices=2)
predictions = trainer.predict(model, dm)

Parameters

• batch – Current batch.

• batch_idx – Index of current batch.

• dataloader_idx – Index of the current dataloader.

Returns

Predicted output

prepare_data() → None

Use this to download and prepare data. Downloading and saving data with multiple processes (distributed settings) will result in corrupted data. Lightning ensures this method is called only within a single process, so you can safely add your downloading logic within.

Warning

DO NOT set state to the model (use setup instead) since this is NOT called on every device

Example:

def prepare_data(self):
    # good
    download_data()
    tokenize()
    etc()

    # bad
    self.split = data_split
    self.some_state = some_other_state()

In a distributed environment, prepare_data can be called in two ways (using prepare_data_per_node)

1. Once per node. This is the default and is only called on LOCAL_RANK=0.

2. Once in total. Only called on GLOBAL_RANK=0.

Example:

# DEFAULT
# called once per node on LOCAL_RANK=0 of that node
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = True


# call on GLOBAL_RANK=0 (great for shared file systems)
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = False

This is called before requesting the dataloaders:

model.prepare_data()
initialize_distributed()
model.setup(stage)
model.train_dataloader()
model.val_dataloader()
model.test_dataloader()
model.predict_dataloader()

print(*args: Any, **kwargs: Any) → None

Prints only from process 0. Use this in any distributed mode to log only once.

Parameters

• *args – The thing to print. The same as for Python’s built-in print function.

• **kwargs – The same as for Python’s built-in print function.

Example:

def forward(self, x):
    self.print(x, 'in forward')

register_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

This function is deprecated in favor of register_full_backward_hook() and the behavior of this function will change in future versions.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_buffer(name: str, tensor: Optional[Tensor], persistent: bool = True) → None

Adds a buffer to the module.

This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the module’s state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module’s state_dict.

Buffers can be accessed as attributes using given names.

Parameters

• name (str) – name of the buffer. The buffer can be accessed from this module using the given name

• tensor (Tensor or None) – buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module’s state_dict.

• persistent (bool) – whether the buffer is part of this module’s state_dict.

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> self.register_buffer('running_mean', torch.zeros(num_features))

register_forward_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output. It should have the following signature:

hook(module, input, output) -> None or modified output

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward() is called.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_forward_pre_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked. It should have the following signature:

hook(module, input) -> None or modified input

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned(unless that value is already a tuple).

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_full_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

The hook will be called every time the gradients with respect to module inputs are computed. The hook should have the following signature:

hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module’s forward function.

Warning

Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_load_state_dict_post_hook(hook)

Registers a post hook to be run after module’s load_state_dict is called.

It should have the following signature::

hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict() with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearning out both missing and unexpected keys will avoid an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_module(name: str, module: Optional[Module]) → None

Alias for add_module().

register_parameter(name: str, param: Optional[Parameter]) → None

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Parameters

• name (str) – name of the parameter. The parameter can be accessed from this module using the given name

• param (Parameter or None) – parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module’s state_dict.

requires_grad_(requires_grad: bool = True) → T

Change if autograd should record operations on parameters in this module.

This method sets the parameters’ requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See Locally disabling gradient computation for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Parameters

requires_grad (bool) – whether autograd should record operations on parameters in this module. Default: True.

Returns

self

Return type

Module

save_hyperparameters(*args: Any, ignore: Optional[Union[Sequence[str], str]] = None, frame: Optional[frame] = None, logger: bool = True) → None

Save arguments to hparams attribute.

Parameters

• args – single object of dict, NameSpace or OmegaConf or string names or arguments from class __init__

• ignore – an argument name or a list of argument names from class __init__ to be ignored

• frame – a frame object. Default is None

• logger – Whether to send the hyperparameters to the logger. Default: True

Example::

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # manually assign arguments
...         self.save_hyperparameters('arg1', 'arg3')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

>>> class AutomaticArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # equivalent automatic
...         self.save_hyperparameters()
...     def forward(self, *args, **kwargs):
...         ...
>>> model = AutomaticArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg2": abc
"arg3": 3.14

>>> class SingleArgModel(HyperparametersMixin):
...     def __init__(self, params):
...         super().__init__()
...         # manually assign single argument
...         self.save_hyperparameters(params)
...     def forward(self, *args, **kwargs):
...         ...
>>> model = SingleArgModel(Namespace(p1=1, p2='abc', p3=3.14))
>>> model.hparams
"p1": 1
"p2": abc
"p3": 3.14

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # pass argument(s) to ignore as a string or in a list
...         self.save_hyperparameters(ignore='arg2')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

set_extra_state(state: Any)

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Parameters

state (dict) – Extra state from the state_dict

setup(stage: str) → None

Called at the beginning of fit (train + validate), validate, test, or predict. This is a good hook when you need to build models dynamically or adjust something about them. This hook is called on every process when using DDP.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

Example:

class LitModel(...):
    def __init__(self):
        self.l1 = None

    def prepare_data(self):
        download_data()
        tokenize()

        # don't do this
        self.something = else

    def setup(self, stage):
        data = load_data(...)
        self.l1 = nn.Linear(28, data.num_classes)

share_memory() → T

See torch.Tensor.share_memory_()

state_dict(*args, destination=None, prefix='', keep_vars=False)

Returns a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

Note

The returned object is a shallow copy. It contains references to the module’s parameters and buffers.

Warning

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

Warning

Please avoid the use of argument destination as it is not designed for end-users.

Parameters

• destination (dict, optional) – If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned. Default: None.

• prefix (str, optional) – a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.

• keep_vars (bool, optional) – by default the Tensor s returned in the state dict are detached from autograd. If it’s set to True, detaching will not be performed. Default: False.

Returns

a dictionary containing a whole state of the module

Return type

dict

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> module.state_dict().keys()
['bias', 'weight']

tbptt_split_batch(batch: Any, split_size: int) → List[Any]

When using truncated backpropagation through time, each batch must be split along the time dimension. Lightning handles this by default, but for custom behavior override this function.

Parameters

• batch – Current batch

• split_size – The size of the split

Returns

List of batch splits. Each split will be passed to training_step() to enable truncated back propagation through time. The default implementation splits root level Tensors and Sequences at dim=1 (i.e. time dim). It assumes that each time dim is the same length.

Examples:

def tbptt_split_batch(self, batch, split_size):
    splits = []
    for t in range(0, time_dims[0], split_size):
        batch_split = []
        for i, x in enumerate(batch):
            if isinstance(x, torch.Tensor):
                split_x = x[:, t:t + split_size]
            elif isinstance(x, collections.abc.Sequence):
                split_x = [None] * len(x)
                for batch_idx in range(len(x)):
                  split_x[batch_idx] = x[batch_idx][t:t + split_size]
            batch_split.append(split_x)
        splits.append(batch_split)
    return splits

Note

Called in the training loop after on_train_batch_start() if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0. Each returned batch split is passed separately to training_step().

teardown(stage: str) → None

Called at the end of fit (train + validate), validate, test, or predict.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

test_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for testing.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• test()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying testing samples.

Example:

def test_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def test_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a test dataset and a test_step(), you don’t need to implement this method.

Note

In the case where you return multiple test dataloaders, the test_step() will have an argument dataloader_idx which matches the order here.

test_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of a test epoch with the output of all test steps.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

outputs – List of outputs you defined in test_step_end(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader

Returns

None

Note

If you didn’t define a test_step(), this won’t be called.

Examples

With a single dataloader:

def test_epoch_end(self, outputs):
    # do something with the outputs of all test batches
    all_test_preds = test_step_outputs.predictions

    some_result = calc_all_results(all_test_preds)
    self.log(some_result)

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each test step for that dataloader.

def test_epoch_end(self, outputs):
    final_value = 0
    for dataloader_outputs in outputs:
        for test_step_out in dataloader_outputs:
            # do something
            final_value += test_step_out

    self.log("final_metric", final_value)

test_step(batch, batch_idx)

Operates on a single batch of data from the test set. In this step you’d normally generate examples or calculate anything of interest such as accuracy.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_id – The index of the dataloader that produced this batch. (only if multiple test dataloaders used).

Returns

Any of.

• Any object or value

• None - Testing will skip to the next batch

# if you have one test dataloader:
def test_step(self, batch, batch_idx):
    ...


# if you have multiple test dataloaders:
def test_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single test dataset
def test_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    test_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'test_loss': loss, 'test_acc': test_acc})

If you pass in multiple test dataloaders, test_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple test dataloaders
def test_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to test you don’t need to implement this method.

Note

When the test_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of the test epoch, the model goes back to training mode and gradients are enabled.

test_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when testing with DP because test_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [test_step(sub_batch) for sub_batch in sub_batches]
test_step_end(step_output)

Parameters

step_output – What you return in test_step() for each batch part.

Returns

None or anything

# WITHOUT test_step_end
# if used in DP, this batch is 1/num_gpus large
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    loss = self.softmax(out)
    self.log("test_loss", loss)


# --------------
# with test_step_end to do softmax over the full batch
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return out


def test_step_end(self, output_results):
    # this out is now the full size of the batch
    all_test_step_outs = output_results.out
    loss = nce_loss(all_test_step_outs)
    self.log("test_loss", loss)

See also

See the Multi GPU Training guide for more details.

to(*args: Any, **kwargs: Any) → Self

See torch.nn.Module.to().

to_empty(*, device: Union[str, device]) → T

Moves the parameters and buffers to the specified device without copying storage.

Parameters

device (torch.device) – The desired device of the parameters and buffers in this module.

Returns

self

Return type

Module

to_onnx(file_path: Union[str, Path], input_sample: Optional[Any] = None, **kwargs: Any) → None

Saves the model in ONNX format.

Parameters

• file_path – The path of the file the onnx model should be saved to.

• input_sample – An input for tracing. Default: None (Use self.example_input_array)

• **kwargs – Will be passed to torch.onnx.export function.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))

>>> with tempfile.NamedTemporaryFile(suffix='.onnx', delete=False) as tmpfile:
...     model = SimpleModel()
...     input_sample = torch.randn((1, 64))
...     model.to_onnx(tmpfile.name, input_sample, export_params=True)
...     os.path.isfile(tmpfile.name)
True

to_torchscript(file_path: Optional[Union[str, Path]] = None, method: Optional[str] = 'script', example_inputs: Optional[Any] = None, **kwargs: Any) → Union[ScriptModule, Dict[str, ScriptModule]]

By default compiles the whole model to a ScriptModule. If you want to use tracing, please provided the argument method='trace' and make sure that either the example_inputs argument is provided, or the model has example_input_array set. If you would like to customize the modules that are scripted you should override this method. In case you want to return multiple modules, we recommend using a dictionary.

Parameters

• file_path – Path where to save the torchscript. Default: None (no file saved).

• method – Whether to use TorchScript’s script or trace method. Default: ‘script’

• example_inputs – An input to be used to do tracing when method is set to ‘trace’. Default: None (uses example_input_array)

• **kwargs – Additional arguments that will be passed to the torch.jit.script() or torch.jit.trace() function.

Note

• Requires the implementation of the forward() method.

• The exported script will be set to evaluation mode.

• It is recommended that you install the latest supported version of PyTorch to use this feature without limitations. See also the torch.jit documentation for supported features.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))
...
>>> model = SimpleModel()
>>> model.to_torchscript(file_path="model.pt")  
>>> os.path.isfile("model.pt")  
>>> torch.jit.save(model.to_torchscript(file_path="model_trace.pt", method='trace', 
...                                     example_inputs=torch.randn(1, 64)))  
>>> os.path.isfile("model_trace.pt")  
True

Returns

This LightningModule as a torchscript, regardless of whether file_path is defined or not.

toggle_optimizer(optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int) → None

Makes sure only the gradients of the current optimizer’s parameters are calculated in the training step to prevent dangling gradients in multiple-optimizer setup.

This is only called automatically when automatic optimization is enabled and multiple optimizers are used. It works with untoggle_optimizer() to make sure param_requires_grad_state is properly reset.

Parameters

• optimizer – The optimizer to toggle.

• optimizer_idx – The index of the optimizer to toggle.

train(mode: bool = True) → T

Sets the module in training mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

Parameters

mode (bool) – whether to set training mode (True) or evaluation mode (False). Default: True.

Returns

self

Return type

Module

train_dataloader() → Union[DataLoader, Sequence[DataLoader], Sequence[Sequence[DataLoader]], Sequence[Dict[str, DataLoader]], Dict[str, DataLoader], Dict[str, Dict[str, DataLoader]], Dict[str, Sequence[DataLoader]]]

Implement one or more PyTorch DataLoaders for training.

Returns

A collection of torch.utils.data.DataLoader specifying training samples. In the case of multiple dataloaders, please see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• fit()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Example:

# single dataloader
def train_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=True, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=True
    )
    return loader

# multiple dataloaders, return as list
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a list of tensors: [batch_mnist, batch_cifar]
    return [mnist_loader, cifar_loader]

# multiple dataloader, return as dict
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a dict of tensors: {'mnist': batch_mnist, 'cifar': batch_cifar}
    return {'mnist': mnist_loader, 'cifar': cifar_loader}

training_epoch_end(outputs: List[Union[Tensor, Dict[str, Any]]]) → None

Called at the end of the training epoch with the outputs of all training steps. Use this in case you need to do something with all the outputs returned by training_step().

# the pseudocode for these calls
train_outs = []
for train_batch in train_data:
    out = training_step(train_batch)
    train_outs.append(out)
training_epoch_end(train_outs)

Parameters

outputs – List of outputs you defined in training_step(). If there are multiple optimizers or when using truncated_bptt_steps > 0, the lists have the dimensions (n_batches, tbptt_steps, n_optimizers). Dimensions of length 1 are squeezed.

Returns

None

Note

If this method is not overridden, this won’t be called.

def training_epoch_end(self, training_step_outputs):
    # do something with all training_step outputs
    for out in training_step_outputs:
        ...

training_step(batch, batch_idx)

Here you compute and return the training loss and some additional metrics for e.g. the progress bar or logger.

Parameters

• batch (Tensor | (Tensor, …) | [Tensor, …]) – The output of your DataLoader. A tensor, tuple or list.

• batch_idx (int) – Integer displaying index of this batch

• optimizer_idx (int) – When using multiple optimizers, this argument will also be present.

• hiddens (Any) – Passed in if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0.

Returns

Any of.

• Tensor - The loss tensor

• dict - A dictionary. Can include any keys, but must include the key 'loss'

• None - Training will skip to the next batch. This is only for automatic optimization.

  This is not supported for multi-GPU, TPU, IPU, or DeepSpeed.

In this step you’d normally do the forward pass and calculate the loss for a batch. You can also do fancier things like multiple forward passes or something model specific.

Example:

def training_step(self, batch, batch_idx):
    x, y, z = batch
    out = self.encoder(x)
    loss = self.loss(out, x)
    return loss

If you define multiple optimizers, this step will be called with an additional optimizer_idx parameter.

# Multiple optimizers (e.g.: GANs)
def training_step(self, batch, batch_idx, optimizer_idx):
    if optimizer_idx == 0:
        # do training_step with encoder
        ...
    if optimizer_idx == 1:
        # do training_step with decoder
        ...

If you add truncated back propagation through time you will also get an additional argument with the hidden states of the previous step.

# Truncated back-propagation through time
def training_step(self, batch, batch_idx, hiddens):
    # hiddens are the hidden states from the previous truncated backprop step
    out, hiddens = self.lstm(data, hiddens)
    loss = ...
    return {"loss": loss, "hiddens": hiddens}

Note

The loss value shown in the progress bar is smoothed (averaged) over the last values, so it differs from the actual loss returned in train/validation step.

Note

When accumulate_grad_batches > 1, the loss returned here will be automatically normalized by accumulate_grad_batches internally.

training_step_end(step_output: Union[Tensor, Dict[str, Any]]) → Union[Tensor, Dict[str, Any]]

Use this when training with dp because training_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [training_step(sub_batch) for sub_batch in sub_batches]
training_step_end(step_output)

Parameters

step_output – What you return in training_step for each batch part.

Returns

Anything

When using the DP strategy, only a portion of the batch is inside the training_step:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)

    # softmax uses only a portion of the batch in the denominator
    loss = self.softmax(out)
    loss = nce_loss(loss)
    return loss

If you wish to do something with all the parts of the batch, then use this method to do it:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return {"pred": out}


def training_step_end(self, training_step_outputs):
    gpu_0_pred = training_step_outputs[0]["pred"]
    gpu_1_pred = training_step_outputs[1]["pred"]
    gpu_n_pred = training_step_outputs[n]["pred"]

    # this softmax now uses the full batch
    loss = nce_loss([gpu_0_pred, gpu_1_pred, gpu_n_pred])
    return loss

See also

See the Multi GPU Training guide for more details.

transfer_batch_to_device(batch: Any, device: device, dataloader_idx: int) → Any

Override this hook if your DataLoader returns tensors wrapped in a custom data structure.

The data types listed below (and any arbitrary nesting of them) are supported out of the box:

• torch.Tensor or anything that implements .to(…)

• list

• dict

• tuple

For anything else, you need to define how the data is moved to the target device (CPU, GPU, TPU, …).

Note

This hook should only transfer the data and not modify it, nor should it move the data to any other device than the one passed in as argument (unless you know what you are doing). To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be transferred to a new device.

• device – The target device as defined in PyTorch.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A reference to the data on the new device.

Example:

def transfer_batch_to_device(self, batch, device, dataloader_idx):
    if isinstance(batch, CustomBatch):
        # move all tensors in your custom data structure to the device
        batch.samples = batch.samples.to(device)
        batch.targets = batch.targets.to(device)
    elif dataloader_idx == 0:
        # skip device transfer for the first dataloader or anything you wish
        pass
    else:
        batch = super().transfer_batch_to_device(data, device, dataloader_idx)
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• move_data_to_device()

• apply_to_collection()

property truncated_bptt_steps: int

Enables Truncated Backpropagation Through Time in the Trainer when set to a positive integer.

It represents the number of times training_step() gets called before backpropagation. If this is > 0, the training_step() receives an additional argument hiddens and is expected to return a hidden state.

type(dst_type: Union[str, dtype]) → Self

See torch.nn.Module.type().

unfreeze() → None

Unfreeze all parameters for training.

model = MyLightningModule(...)
model.unfreeze()

untoggle_optimizer(optimizer_idx: int) → None

Resets the state of required gradients that were toggled with toggle_optimizer().

This is only called automatically when automatic optimization is enabled and multiple optimizers are used.

Parameters

optimizer_idx – The index of the optimizer to untoggle.

val_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for validation.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

It’s recommended that all data downloads and preparation happen in prepare_data().

• fit()

• validate()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying validation samples.

Examples:

def val_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False,
                    transform=transform, download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def val_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a validation dataset and a validation_step(), you don’t need to implement this method.

Note

In the case where you return multiple validation dataloaders, the validation_step() will have an argument dataloader_idx which matches the order here.

validation_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of the validation epoch with the outputs of all validation steps.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

outputs – List of outputs you defined in validation_step(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader.

Returns

None

Note

If you didn’t define a validation_step(), this won’t be called.

Examples

With a single dataloader:

def validation_epoch_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each validation step for that dataloader.

def validation_epoch_end(self, outputs):
    for dataloader_output_result in outputs:
        dataloader_outs = dataloader_output_result.dataloader_i_outputs

    self.log("final_metric", final_value)

validation_step(batch, batch_idx)

Operates on a single batch of data from the validation set. In this step you’d might generate examples or calculate anything of interest like accuracy.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_idx – The index of the dataloader that produced this batch. (only if multiple val dataloaders used)

Returns

• Any object or value

• None - Validation will skip to the next batch

# pseudocode of order
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    if defined("validation_step_end"):
        out = validation_step_end(out)
    val_outs.append(out)
val_outs = validation_epoch_end(val_outs)

# if you have one val dataloader:
def validation_step(self, batch, batch_idx):
    ...


# if you have multiple val dataloaders:
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single validation dataset
def validation_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    val_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'val_loss': loss, 'val_acc': val_acc})

If you pass in multiple val dataloaders, validation_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple validation dataloaders
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to validate you don’t need to implement this method.

Note

When the validation_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of validation, the model goes back to training mode and gradients are enabled.

validation_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when validating with dp because validation_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [validation_step(sub_batch) for sub_batch in sub_batches]
validation_step_end(step_output)

Parameters

step_output – What you return in validation_step() for each batch part.

Returns

None or anything

# WITHOUT validation_step_end
# if used in DP, this batch is 1/num_gpus large
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    loss = self.softmax(out)
    loss = nce_loss(loss)
    self.log("val_loss", loss)


# --------------
# with validation_step_end to do softmax over the full batch
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    return out


def validation_step_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

See also

See the Multi GPU Training guide for more details.

xpu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the XPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

zero_grad(set_to_none: bool = False) → None

Sets gradients of all model parameters to zero. See similar function under torch.optim.Optimizer for more context.

Parameters

set_to_none (bool) – instead of setting to zero, set the grads to None. See torch.optim.Optimizer.zero_grad() for details.

class cca_zoo.deepmodels.BarlowTwins(latent_dims: int, encoders=None, lam=1, **kwargs)

Bases: DCCA

A class used to fit a Barlow Twins model.

References

Zbontar, Jure, et al. “Barlow twins: Self-supervised learning via redundancy reduction.” arXiv preprint arXiv:2103.03230 (2021).

forward(views, **kwargs)

We use the forward model to define the transformation of views to the latent space :param views: batches for each view separated by commas

loss(views, **kwargs)

Loss for model :param views: :param args: :param kwargs: :return:

add_module(name: str, module: Optional[Module]) → None

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Parameters

• name (str) – name of the child module. The child module can be accessed from this module using the given name

• module (Module) – child module to be added to the module.

all_gather(data: Union[Tensor, Dict, List, Tuple], group: Optional[Any] = None, sync_grads: bool = False) → Union[Tensor, Dict, List, Tuple]

Allows users to call self.all_gather() from the LightningModule, thus making the all_gather operation accelerator agnostic. all_gather is a function provided by accelerators to gather a tensor from several distributed processes.

Parameters

• data – int, float, tensor of shape (batch, …), or a (possibly nested) collection thereof.

• group – the process group to gather results from. Defaults to all processes (world)

• sync_grads – flag that allows users to synchronize gradients for the all_gather operation

Returns

A tensor of shape (world_size, batch, …), or if the input was a collection the output will also be a collection with tensors of this shape.

apply(fn: Callable[[Module], None]) → T

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also torch.nn.init).

Parameters

fn (Module -> None) – function to be applied to each submodule

Returns

self

Return type

Module

Example:

>>> @torch.no_grad()
>>> def init_weights(m):
>>>     print(m)
>>>     if type(m) == nn.Linear:
>>>         m.weight.fill_(1.0)
>>>         print(m.weight)
>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

property automatic_optimization: bool

If set to False you are responsible for calling .backward(), .step(), .zero_grad().

backward(loss: Tensor, optimizer: Optional[Steppable], optimizer_idx: Optional[int], *args: Any, **kwargs: Any) → None

Called to perform backward on the loss returned in training_step(). Override this hook with your own implementation if you need to.

Parameters

• loss – The loss tensor returned by training_step(). If gradient accumulation is used, the loss here holds the normalized value (scaled by 1 / accumulation steps).

• optimizer – Current optimizer being used. None if using manual optimization.

• optimizer_idx – Index of the current optimizer being used. None if using manual optimization.

Example:

def backward(self, loss, optimizer, optimizer_idx):
    loss.backward()

bfloat16() → T

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns

self

Return type

Module

buffers(recurse: bool = True) → Iterator[Tensor]

Returns an iterator over module buffers.

Parameters

recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

torch.Tensor – module buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for buf in model.buffers():
>>>     print(type(buf), buf.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

children() → Iterator[Module]

Returns an iterator over immediate children modules.

Yields

Module – a child module

clip_gradients(optimizer: Optimizer, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Handles gradient clipping internally.

Note

Do not override this method. If you want to customize gradient clipping, consider using configure_gradient_clipping() method.

Parameters

• optimizer – Current optimizer being used.

• gradient_clip_val – The value at which to clip gradients.

• gradient_clip_algorithm – The gradient clipping algorithm to use. Pass gradient_clip_algorithm="value" to clip by value, and gradient_clip_algorithm="norm" to clip by norm.

configure_callbacks()

Configure model-specific callbacks. When the model gets attached, e.g., when .fit() or .test() gets called, the list or a callback returned here will be merged with the list of callbacks passed to the Trainer’s callbacks argument. If a callback returned here has the same type as one or several callbacks already present in the Trainer’s callbacks list, it will take priority and replace them. In addition, Lightning will make sure ModelCheckpoint callbacks run last.

Returns

A callback or a list of callbacks which will extend the list of callbacks in the Trainer.

Example:

def configure_callbacks(self):
    early_stop = EarlyStopping(monitor="val_acc", mode="max")
    checkpoint = ModelCheckpoint(monitor="val_loss")
    return [early_stop, checkpoint]

configure_gradient_clipping(optimizer: Optimizer, optimizer_idx: int, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Perform gradient clipping for the optimizer parameters. Called before optimizer_step().

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

• gradient_clip_val – The value at which to clip gradients. By default value passed in Trainer will be available here.

• gradient_clip_algorithm – The gradient clipping algorithm to use. By default value passed in Trainer will be available here.

Example:

# Perform gradient clipping on gradients associated with discriminator (optimizer_idx=1) in GAN
def configure_gradient_clipping(self, optimizer, optimizer_idx, gradient_clip_val, gradient_clip_algorithm):
    if optimizer_idx == 1:
        # Lightning will handle the gradient clipping
        self.clip_gradients(
            optimizer,
            gradient_clip_val=gradient_clip_val,
            gradient_clip_algorithm=gradient_clip_algorithm
        )
    else:
        # implement your own custom logic to clip gradients for generator (optimizer_idx=0)

configure_optimizers()

Collects learnable parameters and configures the optimizer and learning rate scheduler. :returns: two lists containing the optimizer and the scheduler. :rtype: Tuple[List, List]

configure_sharded_model() → None

Hook to create modules in a distributed aware context. This is useful for when using sharded plugins, where we’d like to shard the model instantly, which is useful for extremely large models which can save memory and initialization time.

This hook is called during each of fit/val/test/predict stages in the same process, so ensure that implementation of this hook is idempotent.

cpu() → Self

See torch.nn.Module.cpu().

cuda(device: Optional[Union[device, int]] = None) → Self

Moves all model parameters and buffers to the GPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized.

Parameters

device – If specified, all parameters will be copied to that device. If None, the current CUDA device index will be used.

Returns

self

Return type

Module

property current_epoch: int

The current epoch in the Trainer, or 0 if not attached.

double() → Self

See torch.nn.Module.double().

eval() → T

Sets the module in evaluation mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False).

See Locally disabling gradient computation for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns

self

Return type

Module

property example_input_array: Optional[Union[Tensor, Tuple, Dict]]

The example input array is a specification of what the module can consume in the forward() method. The return type is interpreted as follows:

• Single tensor: It is assumed the model takes a single argument, i.e., model.forward(model.example_input_array)

• Tuple: The input array should be interpreted as a sequence of positional arguments, i.e., model.forward(*model.example_input_array)

• Dict: The input array represents named keyword arguments, i.e., model.forward(**model.example_input_array)

extra_repr() → str

Set the extra representation of the module

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

abstract fit(views: Iterable[ndarray], y=None, **kwargs)

Fits the model to the given data

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• y (None) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

self

Return type

object

fit_transform(views: Iterable[ndarray], **kwargs)

Fits the model to the given data and returns the transformed views

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

transformed_views

Return type

list of numpy arrays

float() → Self

See torch.nn.Module.float().

freeze() → None

Freeze all params for inference.

Example:

model = MyLightningModule(...)
model.freeze()

get_buffer(target: str) → Tensor

Returns the buffer given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The buffer referenced by target

Return type

torch.Tensor

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state() → Any

Returns any extra state to include in the module’s state_dict. Implement this and a corresponding set_extra_state() for your module if you need to store extra state. This function is called when building the module’s state_dict().

Note that extra state should be pickleable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns

Any extra state to store in the module’s state_dict

Return type

object

get_factor_loadings(views: Iterable[ndarray], normalize=True, **kwargs)

Returns the factor loadings for each view

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• normalize (bool, optional) – Whether to normalize the factor loadings. Default is True.

• kwargs (any additional keyword arguments required by the given model) –

Returns

factor_loadings

Return type

list of numpy arrays

get_parameter(target: str) → Parameter

Returns the parameter given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The Parameter referenced by target

Return type

torch.nn.Parameter

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_params(deep=True)

Get parameters for this estimator.

Parameters

deep (bool, default=True) – If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns

params – Parameter names mapped to their values.

Return type

dict

get_submodule(target: str) → Module

Returns the submodule given by target if it exists, otherwise throws an error.

For example, let’s say you have an nn.Module A that looks like this:

A(
    (net_b): Module(
        (net_c): Module(
            (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2))
        )
        (linear): Linear(in_features=100, out_features=200, bias=True)
    )
)

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Parameters

target – The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns

The submodule referenced by target

Return type

torch.nn.Module

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Module

property global_rank: int

The index of the current process across all nodes and devices.

property global_step: int

Total training batches seen across all epochs.

If no Trainer is attached, this propery is 0.

half() → Self

See torch.nn.Module.half().

property hparams: Union[AttributeDict, MutableMapping]

The collection of hyperparameters saved with save_hyperparameters(). It is mutable by the user. For the frozen set of initial hyperparameters, use hparams_initial.

Returns

Mutable hyperparameters dictionary

property hparams_initial: AttributeDict

The collection of hyperparameters saved with save_hyperparameters(). These contents are read-only. Manual updates to the saved hyperparameters can instead be performed through hparams.

Returns

immutable initial hyperparameters

Return type

AttributeDict

ipu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the IPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on IPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

classmethod load_from_checkpoint(checkpoint_path: Union[str, Path, IO], map_location: Optional[Union[device, str, int, Callable[[Union[device, str, int]], Union[device, str, int]], Dict[Union[device, str, int], Union[device, str, int]]]] = None, hparams_file: Optional[Union[str, Path]] = None, strict: bool = True, **kwargs: Any) → Self

Primary way of loading a model from a checkpoint. When Lightning saves a checkpoint it stores the arguments passed to __init__ in the checkpoint under "hyper_parameters".

Any arguments specified through **kwargs will override args stored in "hyper_parameters".

Parameters

• checkpoint_path – Path to checkpoint. This can also be a URL, or file-like object

• map_location – If your checkpoint saved a GPU model and you now load on CPUs or a different number of GPUs, use this to map to the new setup. The behaviour is the same as in torch.load().

• hparams_file –

  Optional path to a .yaml or .csv file with hierarchical structure as in this example:

  drop_prob: 0.2
  dataloader:
      batch_size: 32
  

  You most likely won’t need this since Lightning will always save the hyperparameters to the checkpoint. However, if your checkpoint weights don’t have the hyperparameters saved, use this method to pass in a .yaml file with the hparams you’d like to use. These will be converted into a dict and passed into your LightningModule for use.

  If your model’s hparams argument is Namespace and .yaml file has hierarchical structure, you need to refactor your model to treat hparams as dict.

• strict – Whether to strictly enforce that the keys in checkpoint_path match the keys returned by this module’s state dict.

• **kwargs – Any extra keyword args needed to init the model. Can also be used to override saved hyperparameter values.

Returns

LightningModule instance with loaded weights and hyperparameters (if available).

Note

load_from_checkpoint is a class method. You should use your LightningModule class to call it instead of the LightningModule instance.

Example:

# load weights without mapping ...
model = MyLightningModule.load_from_checkpoint('path/to/checkpoint.ckpt')

# or load weights mapping all weights from GPU 1 to GPU 0 ...
map_location = {'cuda:1':'cuda:0'}
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    map_location=map_location
)

# or load weights and hyperparameters from separate files.
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    hparams_file='/path/to/hparams_file.yaml'
)

# override some of the params with new values
model = MyLightningModule.load_from_checkpoint(
    PATH,
    num_layers=128,
    pretrained_ckpt_path=NEW_PATH,
)

# predict
pretrained_model.eval()
pretrained_model.freeze()
y_hat = pretrained_model(x)

load_state_dict(state_dict: Mapping[str, Any], strict: bool = True)

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Parameters

• state_dict (dict) – a dict containing parameters and persistent buffers.

• strict (bool, optional) – whether to strictly enforce that the keys in state_dict match the keys returned by this module’s state_dict() function. Default: True

Returns

• missing_keys is a list of str containing the missing keys

• unexpected_keys is a list of str containing the unexpected keys

Return type

NamedTuple with missing_keys and unexpected_keys fields

Note

If a parameter or buffer is registered as None and its corresponding key exists in state_dict, load_state_dict() will raise a RuntimeError.

property local_rank: int

The index of the current process within a single node.

log(name: str, value: Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, metric_attribute: Optional[str] = None, rank_zero_only: bool = False) → None

Log a key, value pair.

Example:

self.log('train_loss', loss)

The default behavior per hook is documented here: extensions/logging:Automatic Logging.

Parameters

• name – key to log.

• value – value to log. Can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress bar.

• logger – if True logs to the logger.

• on_step – if True logs at this step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto detach the graph.

• sync_dist – if True, reduces the metric across devices. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the DDP group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple dataloaders). If False, user needs to give unique names for each dataloader to not mix the values.

• batch_size – Current batch_size. This will be directly inferred from the loaded batch, but for some data structures you might need to explicitly provide it.

• metric_attribute – To restore the metric state, Lightning requires the reference of the torchmetrics.Metric in your model. This is found automatically if it is a model attribute.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_dict(dictionary: Mapping[str, Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, rank_zero_only: bool = False) → None

Log a dictionary of values at once.

Example:

values = {'loss': loss, 'acc': acc, ..., 'metric_n': metric_n}
self.log_dict(values)

Parameters

• dictionary – key value pairs. The values can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress base.

• logger – if True logs to the logger.

• on_step – if True logs at this step. None auto-logs for training_step but not validation/test_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. None auto-logs for val/test step but not training_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto-detach the graph

• sync_dist – if True, reduces the metric across GPUs/TPUs. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the ddp group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple). If False, user needs to give unique names for each dataloader to not mix values.

• batch_size – Current batch size. This will be directly inferred from the loaded batch, but some data structures might need to explicitly provide it.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_grad_norm(grad_norm_dict: Dict[str, float]) → None

Override this method to change the default behaviour of log_grad_norm.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

grad_norm_dict – Dictionary containing current grad norm metrics

Example:

# DEFAULT
def log_grad_norm(self, grad_norm_dict):
    self.log_dict(grad_norm_dict, on_step=True, on_epoch=True, prog_bar=False, logger=True)

property logger: Optional[Logger]

Reference to the logger object in the Trainer.

property loggers: List[Logger]

Reference to the list of loggers in the Trainer.

lr_scheduler_step(scheduler: Union[_LRScheduler, ReduceLROnPlateau], optimizer_idx: int, metric: Optional[Any]) → None

Override this method to adjust the default way the Trainer calls each scheduler. By default, Lightning calls step() and as shown in the example for each scheduler based on its interval.

Parameters

• scheduler – Learning rate scheduler.

• optimizer_idx – Index of the optimizer associated with this scheduler.

• metric – Value of the monitor used for schedulers like ReduceLROnPlateau.

Examples:

# DEFAULT
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    if metric is None:
        scheduler.step()
    else:
        scheduler.step(metric)

# Alternative way to update schedulers if it requires an epoch value
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    scheduler.step(epoch=self.current_epoch)

lr_schedulers() → Union[None, List[Union[_LRScheduler, ReduceLROnPlateau]], _LRScheduler, ReduceLROnPlateau]

Returns the learning rate scheduler(s) that are being used during training. Useful for manual optimization.

Returns

A single scheduler, or a list of schedulers in case multiple ones are present, or None if no schedulers were returned in configure_optimizers().

manual_backward(loss: Tensor, *args: Any, **kwargs: Any) → None

Call this directly from your training_step() when doing optimizations manually. By using this, Lightning can ensure that all the proper scaling gets applied when using mixed precision.

See manual optimization for more examples.

Example:

def training_step(...):
    opt = self.optimizers()
    loss = ...
    opt.zero_grad()
    # automatically applies scaling, etc...
    self.manual_backward(loss)
    opt.step()

Parameters

• loss – The tensor on which to compute gradients. Must have a graph attached.

• *args – Additional positional arguments to be forwarded to backward()

• **kwargs – Additional keyword arguments to be forwarded to backward()

modules() → Iterator[Module]

Returns an iterator over all modules in the network.

Yields

Module – a module in the network

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.modules()):
...     print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Tensor]]

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Parameters

• prefix (str) – prefix to prepend to all buffer names.

• recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

(str, torch.Tensor) – Tuple containing the name and buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, buf in self.named_buffers():
>>>    if name in ['running_var']:
>>>        print(buf.size())

named_children() → Iterator[Tuple[str, Module]]

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields

(str, Module) – Tuple containing a name and child module

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, module in model.named_children():
>>>     if name in ['conv4', 'conv5']:
>>>         print(module)

named_modules(memo: Optional[Set[Module]] = None, prefix: str = '', remove_duplicate: bool = True)

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Parameters

• memo – a memo to store the set of modules already added to the result

• prefix – a prefix that will be added to the name of the module

• remove_duplicate – whether to remove the duplicated module instances in the result or not

Yields

(str, Module) – Tuple of name and module

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.named_modules()):
...     print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Parameter]]

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Parameters

• prefix (str) – prefix to prepend to all parameter names.

• recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

(str, Parameter) – Tuple containing the name and parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, param in self.named_parameters():
>>>    if name in ['bias']:
>>>        print(param.size())

on_after_backward() → None

Called after loss.backward() and before optimizers are stepped.

Note

If using native AMP, the gradients will not be unscaled at this point. Use the on_before_optimizer_step if you need the unscaled gradients.

on_after_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch after it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_after_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = gpu_transforms(batch['x'])
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• on_before_batch_transfer()

• transfer_batch_to_device()

on_before_backward(loss: Tensor) → None

Called before loss.backward().

Parameters

loss – Loss divided by number of batches for gradient accumulation and scaled if using native AMP.

on_before_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch before it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_before_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = transforms(batch['x'])
    return batch

See also

• on_after_batch_transfer()

• transfer_batch_to_device()

on_before_optimizer_step(optimizer: Optimizer, optimizer_idx: int) → None

Called before optimizer.step().

If using gradient accumulation, the hook is called once the gradients have been accumulated. See: :paramref:`~pytorch_lightning.trainer.Trainer.accumulate_grad_batches`.

If using native AMP, the loss will be unscaled before calling this hook. See these docs for more information on the scaling of gradients.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

Example:

def on_before_optimizer_step(self, optimizer, optimizer_idx):
    # example to inspect gradient information in tensorboard
    if self.trainer.global_step % 25 == 0:  # don't make the tf file huge
        for k, v in self.named_parameters():
            self.logger.experiment.add_histogram(
                tag=k, values=v.grad, global_step=self.trainer.global_step
            )

on_before_zero_grad(optimizer: Optimizer) → None

Called after training_step() and before optimizer.zero_grad().

Called in the training loop after taking an optimizer step and before zeroing grads. Good place to inspect weight information with weights updated.

This is where it is called:

for optimizer in optimizers:
    out = training_step(...)

    model.on_before_zero_grad(optimizer) # < ---- called here
    optimizer.zero_grad()

    backward()

Parameters

optimizer – The optimizer for which grads should be zeroed.

on_fit_end() → None

Called at the very end of fit.

If on DDP it is called on every process

on_fit_start() → None

Called at the very beginning of fit.

If on DDP it is called on every process

property on_gpu: bool

Returns True if this model is currently located on a GPU.

Useful to set flags around the LightningModule for different CPU vs GPU behavior.

on_load_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning to restore your model. If you saved something with on_save_checkpoint() this is your chance to restore this.

Parameters

checkpoint – Loaded checkpoint

Example:

def on_load_checkpoint(self, checkpoint):
    # 99% of the time you don't need to implement this method
    self.something_cool_i_want_to_save = checkpoint['something_cool_i_want_to_save']

Note

Lightning auto-restores global step, epoch, and train state including amp scaling. There is no need for you to restore anything regarding training.

on_predict_batch_end(outputs: Optional[Any], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop after the batch.

Parameters

• outputs – The outputs of predict_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_end() → None

Called at the end of predicting.

on_predict_epoch_end(results: List[Any]) → None

Called at the end of predicting.

on_predict_epoch_start() → None

Called at the beginning of predicting.

on_predict_model_eval() → None

Sets the model to eval during the predict loop.

on_predict_start() → None

Called at the beginning of predicting.

on_save_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning when saving a checkpoint to give you a chance to store anything else you might want to save.

Parameters

checkpoint – The full checkpoint dictionary before it gets dumped to a file. Implementations of this hook can insert additional data into this dictionary.

Example:

def on_save_checkpoint(self, checkpoint):
    # 99% of use cases you don't need to implement this method
    checkpoint['something_cool_i_want_to_save'] = my_cool_pickable_object

Note

Lightning saves all aspects of training (epoch, global step, etc…) including amp scaling. There is no need for you to store anything about training.

on_test_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop after the batch.

Parameters

• outputs – The outputs of test_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_end() → None

Called at the end of testing.

on_test_epoch_end() → None

Called in the test loop at the very end of the epoch.

on_test_epoch_start() → None

Called in the test loop at the very beginning of the epoch.

on_test_model_eval() → None

Sets the model to eval during the test loop.

on_test_model_train() → None

Sets the model to train during the test loop.

on_test_start() → None

Called at the beginning of testing.

on_train_batch_end(outputs: Union[Tensor, Dict[str, Any]], batch: Any, batch_idx: int) → None

Called in the training loop after the batch.

Parameters

• outputs – The outputs of training_step_end(training_step(x))

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_batch_start(batch: Any, batch_idx: int) → Optional[int]

Called in the training loop before anything happens for that batch.

If you return -1 here, you will skip training for the rest of the current epoch.

Parameters

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_end() → None

Called at the end of training before logger experiment is closed.

on_train_epoch_end() → None

Called in the training loop at the very end of the epoch.

To access all batch outputs at the end of the epoch, either:

1. Implement training_epoch_end in the LightningModule OR

2. Cache data across steps on the attribute(s) of the LightningModule and access them in this hook

on_train_epoch_start() → None

Called in the training loop at the very beginning of the epoch.

on_train_start() → None

Called at the beginning of training after sanity check.

on_validation_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop after the batch.

Parameters

• outputs – The outputs of validation_step_end(validation_step(x))

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_end() → None

Called at the end of validation.

on_validation_epoch_end() → None

Called in the validation loop at the very end of the epoch.

on_validation_epoch_start() → None

Called in the validation loop at the very beginning of the epoch.

on_validation_model_eval() → None

Sets the model to eval during the val loop.

on_validation_model_train() → None

Sets the model to train during the val loop.

on_validation_start() → None

Called at the beginning of validation.

optimizer_step(epoch: int, batch_idx: int, optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int = 0, optimizer_closure: Optional[Callable[[], Any]] = None, on_tpu: bool = False, using_native_amp: bool = False, using_lbfgs: bool = False) → None

Override this method to adjust the default way the Trainer calls each optimizer.

By default, Lightning calls step() and zero_grad() as shown in the example once per optimizer. This method (and zero_grad()) won’t be called during the accumulation phase when Trainer(accumulate_grad_batches != 1). Overriding this hook has no benefit with manual optimization.

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers, this indexes into that list.

• optimizer_closure – The optimizer closure. This closure must be executed as it includes the calls to training_step(), optimizer.zero_grad(), and backward().

• on_tpu – True if TPU backward is required

• using_native_amp – True if using native amp

• using_lbfgs – True if the matching optimizer is torch.optim.LBFGS

Examples:

# DEFAULT
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    optimizer.step(closure=optimizer_closure)

# Alternating schedule for optimizer steps (i.e.: GANs)
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    # update generator opt every step
    if optimizer_idx == 0:
        optimizer.step(closure=optimizer_closure)

    # update discriminator opt every 2 steps
    if optimizer_idx == 1:
        if (batch_idx + 1) % 2 == 0 :
            optimizer.step(closure=optimizer_closure)
        else:
            # call the closure by itself to run `training_step` + `backward` without an optimizer step
            optimizer_closure()

    # ...
    # add as many optimizers as you want

Here’s another example showing how to use this for more advanced things such as learning rate warm-up:

# learning rate warm-up
def optimizer_step(
    self,
    epoch,
    batch_idx,
    optimizer,
    optimizer_idx,
    optimizer_closure,
    on_tpu,
    using_native_amp,
    using_lbfgs,
):
    # update params
    optimizer.step(closure=optimizer_closure)

    # manually warm up lr without a scheduler
    if self.trainer.global_step < 500:
        lr_scale = min(1.0, float(self.trainer.global_step + 1) / 500.0)
        for pg in optimizer.param_groups:
            pg["lr"] = lr_scale * self.learning_rate

optimizer_zero_grad(epoch: int, batch_idx: int, optimizer: Optimizer, optimizer_idx: int) → None

Override this method to change the default behaviour of optimizer.zero_grad().

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers this indexes into that list.

Examples:

# DEFAULT
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad()

# Set gradients to `None` instead of zero to improve performance.
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad(set_to_none=True)

See torch.optim.Optimizer.zero_grad() for the explanation of the above example.

optimizers(use_pl_optimizer: bool = True) → Union[Optimizer, LightningOptimizer, List[Optimizer], List[LightningOptimizer]]

Returns the optimizer(s) that are being used during training. Useful for manual optimization.

Parameters

use_pl_optimizer – If True, will wrap the optimizer(s) in a LightningOptimizer for automatic handling of precision and profiling.

Returns

A single optimizer, or a list of optimizers in case multiple ones are present.

pairwise_correlations(loader: DataLoader, train=False)

Calculates correlation for entire batch from dataloader

Parameters

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

Returns

by default returns the average pairwise correlation in each dimension (for 2 views just the correlation)

parameters(recurse: bool = True) → Iterator[Parameter]

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Parameters

recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

Parameter – module parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for param in model.parameters():
>>>     print(type(param), param.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

predict_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for prediction.

It’s recommended that all data downloads and preparation happen in prepare_data().

• predict()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying prediction samples.

Note

In the case where you return multiple prediction dataloaders, the predict_step() will have an argument dataloader_idx which matches the order here.

predict_step(batch: Any, batch_idx: int, dataloader_idx: int = 0) → Any

Step function called during predict(). By default, it calls forward(). Override to add any processing logic.

The predict_step() is used to scale inference on multi-devices.

To prevent an OOM error, it is possible to use BasePredictionWriter callback to write the predictions to disk or database after each batch or on epoch end.

The BasePredictionWriter should be used while using a spawn based accelerator. This happens for Trainer(strategy="ddp_spawn") or training on 8 TPU cores with Trainer(accelerator="tpu", devices=8) as predictions won’t be returned.

Example

class MyModel(LightningModule):

    def predict_step(self, batch, batch_idx, dataloader_idx=0):
        return self(batch)

dm = ...
model = MyModel()
trainer = Trainer(accelerator="gpu", devices=2)
predictions = trainer.predict(model, dm)

Parameters

• batch – Current batch.

• batch_idx – Index of current batch.

• dataloader_idx – Index of the current dataloader.

Returns

Predicted output

prepare_data() → None

Use this to download and prepare data. Downloading and saving data with multiple processes (distributed settings) will result in corrupted data. Lightning ensures this method is called only within a single process, so you can safely add your downloading logic within.

Warning

DO NOT set state to the model (use setup instead) since this is NOT called on every device

Example:

def prepare_data(self):
    # good
    download_data()
    tokenize()
    etc()

    # bad
    self.split = data_split
    self.some_state = some_other_state()

In a distributed environment, prepare_data can be called in two ways (using prepare_data_per_node)

1. Once per node. This is the default and is only called on LOCAL_RANK=0.

2. Once in total. Only called on GLOBAL_RANK=0.

Example:

# DEFAULT
# called once per node on LOCAL_RANK=0 of that node
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = True


# call on GLOBAL_RANK=0 (great for shared file systems)
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = False

This is called before requesting the dataloaders:

model.prepare_data()
initialize_distributed()
model.setup(stage)
model.train_dataloader()
model.val_dataloader()
model.test_dataloader()
model.predict_dataloader()

print(*args: Any, **kwargs: Any) → None

Prints only from process 0. Use this in any distributed mode to log only once.

Parameters

• *args – The thing to print. The same as for Python’s built-in print function.

• **kwargs – The same as for Python’s built-in print function.

Example:

def forward(self, x):
    self.print(x, 'in forward')

register_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

This function is deprecated in favor of register_full_backward_hook() and the behavior of this function will change in future versions.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_buffer(name: str, tensor: Optional[Tensor], persistent: bool = True) → None

Adds a buffer to the module.

This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the module’s state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module’s state_dict.

Buffers can be accessed as attributes using given names.

Parameters

• name (str) – name of the buffer. The buffer can be accessed from this module using the given name

• tensor (Tensor or None) – buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module’s state_dict.

• persistent (bool) – whether the buffer is part of this module’s state_dict.

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> self.register_buffer('running_mean', torch.zeros(num_features))

register_forward_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output. It should have the following signature:

hook(module, input, output) -> None or modified output

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward() is called.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_forward_pre_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked. It should have the following signature:

hook(module, input) -> None or modified input

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned(unless that value is already a tuple).

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_full_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

The hook will be called every time the gradients with respect to module inputs are computed. The hook should have the following signature:

hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module’s forward function.

Warning

Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_load_state_dict_post_hook(hook)

Registers a post hook to be run after module’s load_state_dict is called.

It should have the following signature::

hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict() with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearning out both missing and unexpected keys will avoid an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_module(name: str, module: Optional[Module]) → None

Alias for add_module().

register_parameter(name: str, param: Optional[Parameter]) → None

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Parameters

• name (str) – name of the parameter. The parameter can be accessed from this module using the given name

• param (Parameter or None) – parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module’s state_dict.

requires_grad_(requires_grad: bool = True) → T

Change if autograd should record operations on parameters in this module.

This method sets the parameters’ requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See Locally disabling gradient computation for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Parameters

requires_grad (bool) – whether autograd should record operations on parameters in this module. Default: True.

Returns

self

Return type

Module

save_hyperparameters(*args: Any, ignore: Optional[Union[Sequence[str], str]] = None, frame: Optional[frame] = None, logger: bool = True) → None

Save arguments to hparams attribute.

Parameters

• args – single object of dict, NameSpace or OmegaConf or string names or arguments from class __init__

• ignore – an argument name or a list of argument names from class __init__ to be ignored

• frame – a frame object. Default is None

• logger – Whether to send the hyperparameters to the logger. Default: True

Example::

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # manually assign arguments
...         self.save_hyperparameters('arg1', 'arg3')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

>>> class AutomaticArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # equivalent automatic
...         self.save_hyperparameters()
...     def forward(self, *args, **kwargs):
...         ...
>>> model = AutomaticArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg2": abc
"arg3": 3.14

>>> class SingleArgModel(HyperparametersMixin):
...     def __init__(self, params):
...         super().__init__()
...         # manually assign single argument
...         self.save_hyperparameters(params)
...     def forward(self, *args, **kwargs):
...         ...
>>> model = SingleArgModel(Namespace(p1=1, p2='abc', p3=3.14))
>>> model.hparams
"p1": 1
"p2": abc
"p3": 3.14

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # pass argument(s) to ignore as a string or in a list
...         self.save_hyperparameters(ignore='arg2')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

score(loader: DataLoader, **kwargs)

Returns average correlation in each dimension (averages over all pairs for multiview)

Parameters

• **kwargs –

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

set_extra_state(state: Any)

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Parameters

state (dict) – Extra state from the state_dict

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters

**params (dict) – Estimator parameters.

Returns

self – Estimator instance.

Return type

estimator instance

setup(stage: str) → None

Called at the beginning of fit (train + validate), validate, test, or predict. This is a good hook when you need to build models dynamically or adjust something about them. This hook is called on every process when using DDP.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

Example:

class LitModel(...):
    def __init__(self):
        self.l1 = None

    def prepare_data(self):
        download_data()
        tokenize()

        # don't do this
        self.something = else

    def setup(self, stage):
        data = load_data(...)
        self.l1 = nn.Linear(28, data.num_classes)

share_memory() → T

See torch.Tensor.share_memory_()

state_dict(*args, destination=None, prefix='', keep_vars=False)

Returns a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

Note

The returned object is a shallow copy. It contains references to the module’s parameters and buffers.

Warning

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

Warning

Please avoid the use of argument destination as it is not designed for end-users.

Parameters

• destination (dict, optional) – If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned. Default: None.

• prefix (str, optional) – a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.

• keep_vars (bool, optional) – by default the Tensor s returned in the state dict are detached from autograd. If it’s set to True, detaching will not be performed. Default: False.

Returns

a dictionary containing a whole state of the module

Return type

dict

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> module.state_dict().keys()
['bias', 'weight']

tbptt_split_batch(batch: Any, split_size: int) → List[Any]

When using truncated backpropagation through time, each batch must be split along the time dimension. Lightning handles this by default, but for custom behavior override this function.

Parameters

• batch – Current batch

• split_size – The size of the split

Returns

List of batch splits. Each split will be passed to training_step() to enable truncated back propagation through time. The default implementation splits root level Tensors and Sequences at dim=1 (i.e. time dim). It assumes that each time dim is the same length.

Examples:

def tbptt_split_batch(self, batch, split_size):
    splits = []
    for t in range(0, time_dims[0], split_size):
        batch_split = []
        for i, x in enumerate(batch):
            if isinstance(x, torch.Tensor):
                split_x = x[:, t:t + split_size]
            elif isinstance(x, collections.abc.Sequence):
                split_x = [None] * len(x)
                for batch_idx in range(len(x)):
                  split_x[batch_idx] = x[batch_idx][t:t + split_size]
            batch_split.append(split_x)
        splits.append(batch_split)
    return splits

Note

Called in the training loop after on_train_batch_start() if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0. Each returned batch split is passed separately to training_step().

teardown(stage: str) → None

Called at the end of fit (train + validate), validate, test, or predict.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

test_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for testing.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• test()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying testing samples.

Example:

def test_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def test_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a test dataset and a test_step(), you don’t need to implement this method.

Note

In the case where you return multiple test dataloaders, the test_step() will have an argument dataloader_idx which matches the order here.

test_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of a test epoch with the output of all test steps.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

outputs – List of outputs you defined in test_step_end(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader

Returns

None

Note

If you didn’t define a test_step(), this won’t be called.

Examples

With a single dataloader:

def test_epoch_end(self, outputs):
    # do something with the outputs of all test batches
    all_test_preds = test_step_outputs.predictions

    some_result = calc_all_results(all_test_preds)
    self.log(some_result)

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each test step for that dataloader.

def test_epoch_end(self, outputs):
    final_value = 0
    for dataloader_outputs in outputs:
        for test_step_out in dataloader_outputs:
            # do something
            final_value += test_step_out

    self.log("final_metric", final_value)

test_step(batch, batch_idx)

Operates on a single batch of data from the test set. In this step you’d normally generate examples or calculate anything of interest such as accuracy.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_id – The index of the dataloader that produced this batch. (only if multiple test dataloaders used).

Returns

Any of.

• Any object or value

• None - Testing will skip to the next batch

# if you have one test dataloader:
def test_step(self, batch, batch_idx):
    ...


# if you have multiple test dataloaders:
def test_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single test dataset
def test_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    test_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'test_loss': loss, 'test_acc': test_acc})

If you pass in multiple test dataloaders, test_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple test dataloaders
def test_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to test you don’t need to implement this method.

Note

When the test_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of the test epoch, the model goes back to training mode and gradients are enabled.

test_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when testing with DP because test_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [test_step(sub_batch) for sub_batch in sub_batches]
test_step_end(step_output)

Parameters

step_output – What you return in test_step() for each batch part.

Returns

None or anything

# WITHOUT test_step_end
# if used in DP, this batch is 1/num_gpus large
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    loss = self.softmax(out)
    self.log("test_loss", loss)


# --------------
# with test_step_end to do softmax over the full batch
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return out


def test_step_end(self, output_results):
    # this out is now the full size of the batch
    all_test_step_outs = output_results.out
    loss = nce_loss(all_test_step_outs)
    self.log("test_loss", loss)

See also

See the Multi GPU Training guide for more details.

to(*args: Any, **kwargs: Any) → Self

See torch.nn.Module.to().

to_empty(*, device: Union[str, device]) → T

Moves the parameters and buffers to the specified device without copying storage.

Parameters

device (torch.device) – The desired device of the parameters and buffers in this module.

Returns

self

Return type

Module

to_onnx(file_path: Union[str, Path], input_sample: Optional[Any] = None, **kwargs: Any) → None

Saves the model in ONNX format.

Parameters

• file_path – The path of the file the onnx model should be saved to.

• input_sample – An input for tracing. Default: None (Use self.example_input_array)

• **kwargs – Will be passed to torch.onnx.export function.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))

>>> with tempfile.NamedTemporaryFile(suffix='.onnx', delete=False) as tmpfile:
...     model = SimpleModel()
...     input_sample = torch.randn((1, 64))
...     model.to_onnx(tmpfile.name, input_sample, export_params=True)
...     os.path.isfile(tmpfile.name)
True

to_torchscript(file_path: Optional[Union[str, Path]] = None, method: Optional[str] = 'script', example_inputs: Optional[Any] = None, **kwargs: Any) → Union[ScriptModule, Dict[str, ScriptModule]]

By default compiles the whole model to a ScriptModule. If you want to use tracing, please provided the argument method='trace' and make sure that either the example_inputs argument is provided, or the model has example_input_array set. If you would like to customize the modules that are scripted you should override this method. In case you want to return multiple modules, we recommend using a dictionary.

Parameters

• file_path – Path where to save the torchscript. Default: None (no file saved).

• method – Whether to use TorchScript’s script or trace method. Default: ‘script’

• example_inputs – An input to be used to do tracing when method is set to ‘trace’. Default: None (uses example_input_array)

• **kwargs – Additional arguments that will be passed to the torch.jit.script() or torch.jit.trace() function.

Note

• Requires the implementation of the forward() method.

• The exported script will be set to evaluation mode.

• It is recommended that you install the latest supported version of PyTorch to use this feature without limitations. See also the torch.jit documentation for supported features.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))
...
>>> model = SimpleModel()
>>> model.to_torchscript(file_path="model.pt")  
>>> os.path.isfile("model.pt")  
>>> torch.jit.save(model.to_torchscript(file_path="model_trace.pt", method='trace', 
...                                     example_inputs=torch.randn(1, 64)))  
>>> os.path.isfile("model_trace.pt")  
True

Returns

This LightningModule as a torchscript, regardless of whether file_path is defined or not.

toggle_optimizer(optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int) → None

Makes sure only the gradients of the current optimizer’s parameters are calculated in the training step to prevent dangling gradients in multiple-optimizer setup.

This is only called automatically when automatic optimization is enabled and multiple optimizers are used. It works with untoggle_optimizer() to make sure param_requires_grad_state is properly reset.

Parameters

• optimizer – The optimizer to toggle.

• optimizer_idx – The index of the optimizer to toggle.

train(mode: bool = True) → T

Sets the module in training mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

Parameters

mode (bool) – whether to set training mode (True) or evaluation mode (False). Default: True.

Returns

self

Return type

Module

train_dataloader() → Union[DataLoader, Sequence[DataLoader], Sequence[Sequence[DataLoader]], Sequence[Dict[str, DataLoader]], Dict[str, DataLoader], Dict[str, Dict[str, DataLoader]], Dict[str, Sequence[DataLoader]]]

Implement one or more PyTorch DataLoaders for training.

Returns

A collection of torch.utils.data.DataLoader specifying training samples. In the case of multiple dataloaders, please see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• fit()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Example:

# single dataloader
def train_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=True, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=True
    )
    return loader

# multiple dataloaders, return as list
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a list of tensors: [batch_mnist, batch_cifar]
    return [mnist_loader, cifar_loader]

# multiple dataloader, return as dict
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a dict of tensors: {'mnist': batch_mnist, 'cifar': batch_cifar}
    return {'mnist': mnist_loader, 'cifar': cifar_loader}

training_epoch_end(outputs: List[Union[Tensor, Dict[str, Any]]]) → None

Called at the end of the training epoch with the outputs of all training steps. Use this in case you need to do something with all the outputs returned by training_step().

# the pseudocode for these calls
train_outs = []
for train_batch in train_data:
    out = training_step(train_batch)
    train_outs.append(out)
training_epoch_end(train_outs)

Parameters

outputs – List of outputs you defined in training_step(). If there are multiple optimizers or when using truncated_bptt_steps > 0, the lists have the dimensions (n_batches, tbptt_steps, n_optimizers). Dimensions of length 1 are squeezed.

Returns

None

Note

If this method is not overridden, this won’t be called.

def training_epoch_end(self, training_step_outputs):
    # do something with all training_step outputs
    for out in training_step_outputs:
        ...

training_step(batch, batch_idx)

Here you compute and return the training loss and some additional metrics for e.g. the progress bar or logger.

Parameters

• batch (Tensor | (Tensor, …) | [Tensor, …]) – The output of your DataLoader. A tensor, tuple or list.

• batch_idx (int) – Integer displaying index of this batch

• optimizer_idx (int) – When using multiple optimizers, this argument will also be present.

• hiddens (Any) – Passed in if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0.

Returns

Any of.

• Tensor - The loss tensor

• dict - A dictionary. Can include any keys, but must include the key 'loss'

• None - Training will skip to the next batch. This is only for automatic optimization.

  This is not supported for multi-GPU, TPU, IPU, or DeepSpeed.

In this step you’d normally do the forward pass and calculate the loss for a batch. You can also do fancier things like multiple forward passes or something model specific.

Example:

def training_step(self, batch, batch_idx):
    x, y, z = batch
    out = self.encoder(x)
    loss = self.loss(out, x)
    return loss

If you define multiple optimizers, this step will be called with an additional optimizer_idx parameter.

# Multiple optimizers (e.g.: GANs)
def training_step(self, batch, batch_idx, optimizer_idx):
    if optimizer_idx == 0:
        # do training_step with encoder
        ...
    if optimizer_idx == 1:
        # do training_step with decoder
        ...

If you add truncated back propagation through time you will also get an additional argument with the hidden states of the previous step.

# Truncated back-propagation through time
def training_step(self, batch, batch_idx, hiddens):
    # hiddens are the hidden states from the previous truncated backprop step
    out, hiddens = self.lstm(data, hiddens)
    loss = ...
    return {"loss": loss, "hiddens": hiddens}

Note

The loss value shown in the progress bar is smoothed (averaged) over the last values, so it differs from the actual loss returned in train/validation step.

Note

When accumulate_grad_batches > 1, the loss returned here will be automatically normalized by accumulate_grad_batches internally.

training_step_end(step_output: Union[Tensor, Dict[str, Any]]) → Union[Tensor, Dict[str, Any]]

Use this when training with dp because training_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [training_step(sub_batch) for sub_batch in sub_batches]
training_step_end(step_output)

Parameters

step_output – What you return in training_step for each batch part.

Returns

Anything

When using the DP strategy, only a portion of the batch is inside the training_step:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)

    # softmax uses only a portion of the batch in the denominator
    loss = self.softmax(out)
    loss = nce_loss(loss)
    return loss

If you wish to do something with all the parts of the batch, then use this method to do it:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return {"pred": out}


def training_step_end(self, training_step_outputs):
    gpu_0_pred = training_step_outputs[0]["pred"]
    gpu_1_pred = training_step_outputs[1]["pred"]
    gpu_n_pred = training_step_outputs[n]["pred"]

    # this softmax now uses the full batch
    loss = nce_loss([gpu_0_pred, gpu_1_pred, gpu_n_pred])
    return loss

See also

See the Multi GPU Training guide for more details.

transfer_batch_to_device(batch: Any, device: device, dataloader_idx: int) → Any

Override this hook if your DataLoader returns tensors wrapped in a custom data structure.

The data types listed below (and any arbitrary nesting of them) are supported out of the box:

• torch.Tensor or anything that implements .to(…)

• list

• dict

• tuple

For anything else, you need to define how the data is moved to the target device (CPU, GPU, TPU, …).

Note

This hook should only transfer the data and not modify it, nor should it move the data to any other device than the one passed in as argument (unless you know what you are doing). To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be transferred to a new device.

• device – The target device as defined in PyTorch.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A reference to the data on the new device.

Example:

def transfer_batch_to_device(self, batch, device, dataloader_idx):
    if isinstance(batch, CustomBatch):
        # move all tensors in your custom data structure to the device
        batch.samples = batch.samples.to(device)
        batch.targets = batch.targets.to(device)
    elif dataloader_idx == 0:
        # skip device transfer for the first dataloader or anything you wish
        pass
    else:
        batch = super().transfer_batch_to_device(data, device, dataloader_idx)
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• move_data_to_device()

• apply_to_collection()

transform(loader: DataLoader)

Parameters

loader – a dataloader that matches the structure of that used for training

Returns

transformed views

property truncated_bptt_steps: int

Enables Truncated Backpropagation Through Time in the Trainer when set to a positive integer.

It represents the number of times training_step() gets called before backpropagation. If this is > 0, the training_step() receives an additional argument hiddens and is expected to return a hidden state.

type(dst_type: Union[str, dtype]) → Self

See torch.nn.Module.type().

unfreeze() → None

Unfreeze all parameters for training.

model = MyLightningModule(...)
model.unfreeze()

untoggle_optimizer(optimizer_idx: int) → None

Resets the state of required gradients that were toggled with toggle_optimizer().

This is only called automatically when automatic optimization is enabled and multiple optimizers are used.

Parameters

optimizer_idx – The index of the optimizer to untoggle.

val_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for validation.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

It’s recommended that all data downloads and preparation happen in prepare_data().

• fit()

• validate()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying validation samples.

Examples:

def val_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False,
                    transform=transform, download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def val_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a validation dataset and a validation_step(), you don’t need to implement this method.

Note

In the case where you return multiple validation dataloaders, the validation_step() will have an argument dataloader_idx which matches the order here.

validation_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of the validation epoch with the outputs of all validation steps.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

outputs – List of outputs you defined in validation_step(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader.

Returns

None

Note

If you didn’t define a validation_step(), this won’t be called.

Examples

With a single dataloader:

def validation_epoch_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each validation step for that dataloader.

def validation_epoch_end(self, outputs):
    for dataloader_output_result in outputs:
        dataloader_outs = dataloader_output_result.dataloader_i_outputs

    self.log("final_metric", final_value)

validation_step(batch, batch_idx)

Operates on a single batch of data from the validation set. In this step you’d might generate examples or calculate anything of interest like accuracy.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_idx – The index of the dataloader that produced this batch. (only if multiple val dataloaders used)

Returns

• Any object or value

• None - Validation will skip to the next batch

# pseudocode of order
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    if defined("validation_step_end"):
        out = validation_step_end(out)
    val_outs.append(out)
val_outs = validation_epoch_end(val_outs)

# if you have one val dataloader:
def validation_step(self, batch, batch_idx):
    ...


# if you have multiple val dataloaders:
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single validation dataset
def validation_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    val_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'val_loss': loss, 'val_acc': val_acc})

If you pass in multiple val dataloaders, validation_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple validation dataloaders
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to validate you don’t need to implement this method.

Note

When the validation_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of validation, the model goes back to training mode and gradients are enabled.

validation_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when validating with dp because validation_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [validation_step(sub_batch) for sub_batch in sub_batches]
validation_step_end(step_output)

Parameters

step_output – What you return in validation_step() for each batch part.

Returns

None or anything

# WITHOUT validation_step_end
# if used in DP, this batch is 1/num_gpus large
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    loss = self.softmax(out)
    loss = nce_loss(loss)
    self.log("val_loss", loss)


# --------------
# with validation_step_end to do softmax over the full batch
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    return out


def validation_step_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

See also

See the Multi GPU Training guide for more details.

xpu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the XPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

zero_grad(set_to_none: bool = False) → None

Sets gradients of all model parameters to zero. See similar function under torch.optim.Optimizer for more context.

Parameters

set_to_none (bool) – instead of setting to zero, set the grads to None. See torch.optim.Optimizer.zero_grad() for details.

class cca_zoo.deepmodels.DTCCA(latent_dims: int, encoders=None, r: float = 0, eps: float = 1e-05, **kwargs)

Bases: DCCA

A class used to fit a DTCCA model.

Is just a thin wrapper round DCCA with the DTCCA objective and a TCCA post-processing

References

Wong, Hok Shing, et al. “Deep Tensor CCA for Multi-view Learning.” IEEE Transactions on Big Data (2021).

add_module(name: str, module: Optional[Module]) → None

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Parameters

• name (str) – name of the child module. The child module can be accessed from this module using the given name

• module (Module) – child module to be added to the module.

all_gather(data: Union[Tensor, Dict, List, Tuple], group: Optional[Any] = None, sync_grads: bool = False) → Union[Tensor, Dict, List, Tuple]

Allows users to call self.all_gather() from the LightningModule, thus making the all_gather operation accelerator agnostic. all_gather is a function provided by accelerators to gather a tensor from several distributed processes.

Parameters

• data – int, float, tensor of shape (batch, …), or a (possibly nested) collection thereof.

• group – the process group to gather results from. Defaults to all processes (world)

• sync_grads – flag that allows users to synchronize gradients for the all_gather operation

Returns

A tensor of shape (world_size, batch, …), or if the input was a collection the output will also be a collection with tensors of this shape.

apply(fn: Callable[[Module], None]) → T

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also torch.nn.init).

Parameters

fn (Module -> None) – function to be applied to each submodule

Returns

self

Return type

Module

Example:

>>> @torch.no_grad()
>>> def init_weights(m):
>>>     print(m)
>>>     if type(m) == nn.Linear:
>>>         m.weight.fill_(1.0)
>>>         print(m.weight)
>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

property automatic_optimization: bool

If set to False you are responsible for calling .backward(), .step(), .zero_grad().

backward(loss: Tensor, optimizer: Optional[Steppable], optimizer_idx: Optional[int], *args: Any, **kwargs: Any) → None

Called to perform backward on the loss returned in training_step(). Override this hook with your own implementation if you need to.

Parameters

• loss – The loss tensor returned by training_step(). If gradient accumulation is used, the loss here holds the normalized value (scaled by 1 / accumulation steps).

• optimizer – Current optimizer being used. None if using manual optimization.

• optimizer_idx – Index of the current optimizer being used. None if using manual optimization.

Example:

def backward(self, loss, optimizer, optimizer_idx):
    loss.backward()

bfloat16() → T

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns

self

Return type

Module

buffers(recurse: bool = True) → Iterator[Tensor]

Returns an iterator over module buffers.

Parameters

recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

torch.Tensor – module buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for buf in model.buffers():
>>>     print(type(buf), buf.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

children() → Iterator[Module]

Returns an iterator over immediate children modules.

Yields

Module – a child module

clip_gradients(optimizer: Optimizer, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Handles gradient clipping internally.

Note

Do not override this method. If you want to customize gradient clipping, consider using configure_gradient_clipping() method.

Parameters

• optimizer – Current optimizer being used.

• gradient_clip_val – The value at which to clip gradients.

• gradient_clip_algorithm – The gradient clipping algorithm to use. Pass gradient_clip_algorithm="value" to clip by value, and gradient_clip_algorithm="norm" to clip by norm.

configure_callbacks()

Configure model-specific callbacks. When the model gets attached, e.g., when .fit() or .test() gets called, the list or a callback returned here will be merged with the list of callbacks passed to the Trainer’s callbacks argument. If a callback returned here has the same type as one or several callbacks already present in the Trainer’s callbacks list, it will take priority and replace them. In addition, Lightning will make sure ModelCheckpoint callbacks run last.

Returns

A callback or a list of callbacks which will extend the list of callbacks in the Trainer.

Example:

def configure_callbacks(self):
    early_stop = EarlyStopping(monitor="val_acc", mode="max")
    checkpoint = ModelCheckpoint(monitor="val_loss")
    return [early_stop, checkpoint]

configure_gradient_clipping(optimizer: Optimizer, optimizer_idx: int, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Perform gradient clipping for the optimizer parameters. Called before optimizer_step().

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

• gradient_clip_val – The value at which to clip gradients. By default value passed in Trainer will be available here.

• gradient_clip_algorithm – The gradient clipping algorithm to use. By default value passed in Trainer will be available here.

Example:

# Perform gradient clipping on gradients associated with discriminator (optimizer_idx=1) in GAN
def configure_gradient_clipping(self, optimizer, optimizer_idx, gradient_clip_val, gradient_clip_algorithm):
    if optimizer_idx == 1:
        # Lightning will handle the gradient clipping
        self.clip_gradients(
            optimizer,
            gradient_clip_val=gradient_clip_val,
            gradient_clip_algorithm=gradient_clip_algorithm
        )
    else:
        # implement your own custom logic to clip gradients for generator (optimizer_idx=0)

configure_optimizers()

Collects learnable parameters and configures the optimizer and learning rate scheduler. :returns: two lists containing the optimizer and the scheduler. :rtype: Tuple[List, List]

configure_sharded_model() → None

Hook to create modules in a distributed aware context. This is useful for when using sharded plugins, where we’d like to shard the model instantly, which is useful for extremely large models which can save memory and initialization time.

This hook is called during each of fit/val/test/predict stages in the same process, so ensure that implementation of this hook is idempotent.

cpu() → Self

See torch.nn.Module.cpu().

cuda(device: Optional[Union[device, int]] = None) → Self

Moves all model parameters and buffers to the GPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized.

Parameters

device – If specified, all parameters will be copied to that device. If None, the current CUDA device index will be used.

Returns

self

Return type

Module

property current_epoch: int

The current epoch in the Trainer, or 0 if not attached.

double() → Self

See torch.nn.Module.double().

eval() → T

Sets the module in evaluation mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False).

See Locally disabling gradient computation for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns

self

Return type

Module

property example_input_array: Optional[Union[Tensor, Tuple, Dict]]

The example input array is a specification of what the module can consume in the forward() method. The return type is interpreted as follows:

• Single tensor: It is assumed the model takes a single argument, i.e., model.forward(model.example_input_array)

• Tuple: The input array should be interpreted as a sequence of positional arguments, i.e., model.forward(*model.example_input_array)

• Dict: The input array represents named keyword arguments, i.e., model.forward(**model.example_input_array)

extra_repr() → str

Set the extra representation of the module

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

abstract fit(views: Iterable[ndarray], y=None, **kwargs)

Fits the model to the given data

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• y (None) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

self

Return type

object

fit_transform(views: Iterable[ndarray], **kwargs)

Fits the model to the given data and returns the transformed views

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

transformed_views

Return type

list of numpy arrays

float() → Self

See torch.nn.Module.float().

forward(views, **kwargs)

We use the forward model to define the transformation of views to the latent space :param views: batches for each view separated by commas

freeze() → None

Freeze all params for inference.

Example:

model = MyLightningModule(...)
model.freeze()

get_buffer(target: str) → Tensor

Returns the buffer given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The buffer referenced by target

Return type

torch.Tensor

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state() → Any

Returns any extra state to include in the module’s state_dict. Implement this and a corresponding set_extra_state() for your module if you need to store extra state. This function is called when building the module’s state_dict().

Note that extra state should be pickleable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns

Any extra state to store in the module’s state_dict

Return type

object

get_factor_loadings(views: Iterable[ndarray], normalize=True, **kwargs)

Returns the factor loadings for each view

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• normalize (bool, optional) – Whether to normalize the factor loadings. Default is True.

• kwargs (any additional keyword arguments required by the given model) –

Returns

factor_loadings

Return type

list of numpy arrays

get_parameter(target: str) → Parameter

Returns the parameter given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The Parameter referenced by target

Return type

torch.nn.Parameter

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_params(deep=True)

Get parameters for this estimator.

Parameters

deep (bool, default=True) – If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns

params – Parameter names mapped to their values.

Return type

dict

get_submodule(target: str) → Module

Returns the submodule given by target if it exists, otherwise throws an error.

For example, let’s say you have an nn.Module A that looks like this:

A(
    (net_b): Module(
        (net_c): Module(
            (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2))
        )
        (linear): Linear(in_features=100, out_features=200, bias=True)
    )
)

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Parameters

target – The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns

The submodule referenced by target

Return type

torch.nn.Module

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Module

property global_rank: int

The index of the current process across all nodes and devices.

property global_step: int

Total training batches seen across all epochs.

If no Trainer is attached, this propery is 0.

half() → Self

See torch.nn.Module.half().

property hparams: Union[AttributeDict, MutableMapping]

The collection of hyperparameters saved with save_hyperparameters(). It is mutable by the user. For the frozen set of initial hyperparameters, use hparams_initial.

Returns

Mutable hyperparameters dictionary

property hparams_initial: AttributeDict

The collection of hyperparameters saved with save_hyperparameters(). These contents are read-only. Manual updates to the saved hyperparameters can instead be performed through hparams.

Returns

immutable initial hyperparameters

Return type

AttributeDict

ipu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the IPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on IPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

classmethod load_from_checkpoint(checkpoint_path: Union[str, Path, IO], map_location: Optional[Union[device, str, int, Callable[[Union[device, str, int]], Union[device, str, int]], Dict[Union[device, str, int], Union[device, str, int]]]] = None, hparams_file: Optional[Union[str, Path]] = None, strict: bool = True, **kwargs: Any) → Self

Primary way of loading a model from a checkpoint. When Lightning saves a checkpoint it stores the arguments passed to __init__ in the checkpoint under "hyper_parameters".

Any arguments specified through **kwargs will override args stored in "hyper_parameters".

Parameters

• checkpoint_path – Path to checkpoint. This can also be a URL, or file-like object

• map_location – If your checkpoint saved a GPU model and you now load on CPUs or a different number of GPUs, use this to map to the new setup. The behaviour is the same as in torch.load().

• hparams_file –

  Optional path to a .yaml or .csv file with hierarchical structure as in this example:

  drop_prob: 0.2
  dataloader:
      batch_size: 32
  

  You most likely won’t need this since Lightning will always save the hyperparameters to the checkpoint. However, if your checkpoint weights don’t have the hyperparameters saved, use this method to pass in a .yaml file with the hparams you’d like to use. These will be converted into a dict and passed into your LightningModule for use.

  If your model’s hparams argument is Namespace and .yaml file has hierarchical structure, you need to refactor your model to treat hparams as dict.

• strict – Whether to strictly enforce that the keys in checkpoint_path match the keys returned by this module’s state dict.

• **kwargs – Any extra keyword args needed to init the model. Can also be used to override saved hyperparameter values.

Returns

LightningModule instance with loaded weights and hyperparameters (if available).

Note

load_from_checkpoint is a class method. You should use your LightningModule class to call it instead of the LightningModule instance.

Example:

# load weights without mapping ...
model = MyLightningModule.load_from_checkpoint('path/to/checkpoint.ckpt')

# or load weights mapping all weights from GPU 1 to GPU 0 ...
map_location = {'cuda:1':'cuda:0'}
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    map_location=map_location
)

# or load weights and hyperparameters from separate files.
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    hparams_file='/path/to/hparams_file.yaml'
)

# override some of the params with new values
model = MyLightningModule.load_from_checkpoint(
    PATH,
    num_layers=128,
    pretrained_ckpt_path=NEW_PATH,
)

# predict
pretrained_model.eval()
pretrained_model.freeze()
y_hat = pretrained_model(x)

load_state_dict(state_dict: Mapping[str, Any], strict: bool = True)

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Parameters

• state_dict (dict) – a dict containing parameters and persistent buffers.

• strict (bool, optional) – whether to strictly enforce that the keys in state_dict match the keys returned by this module’s state_dict() function. Default: True

Returns

• missing_keys is a list of str containing the missing keys

• unexpected_keys is a list of str containing the unexpected keys

Return type

NamedTuple with missing_keys and unexpected_keys fields

Note

If a parameter or buffer is registered as None and its corresponding key exists in state_dict, load_state_dict() will raise a RuntimeError.

property local_rank: int

The index of the current process within a single node.

log(name: str, value: Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, metric_attribute: Optional[str] = None, rank_zero_only: bool = False) → None

Log a key, value pair.

Example:

self.log('train_loss', loss)

The default behavior per hook is documented here: extensions/logging:Automatic Logging.

Parameters

• name – key to log.

• value – value to log. Can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress bar.

• logger – if True logs to the logger.

• on_step – if True logs at this step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto detach the graph.

• sync_dist – if True, reduces the metric across devices. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the DDP group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple dataloaders). If False, user needs to give unique names for each dataloader to not mix the values.

• batch_size – Current batch_size. This will be directly inferred from the loaded batch, but for some data structures you might need to explicitly provide it.

• metric_attribute – To restore the metric state, Lightning requires the reference of the torchmetrics.Metric in your model. This is found automatically if it is a model attribute.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_dict(dictionary: Mapping[str, Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, rank_zero_only: bool = False) → None

Log a dictionary of values at once.

Example:

values = {'loss': loss, 'acc': acc, ..., 'metric_n': metric_n}
self.log_dict(values)

Parameters

• dictionary – key value pairs. The values can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress base.

• logger – if True logs to the logger.

• on_step – if True logs at this step. None auto-logs for training_step but not validation/test_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. None auto-logs for val/test step but not training_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto-detach the graph

• sync_dist – if True, reduces the metric across GPUs/TPUs. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the ddp group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple). If False, user needs to give unique names for each dataloader to not mix values.

• batch_size – Current batch size. This will be directly inferred from the loaded batch, but some data structures might need to explicitly provide it.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_grad_norm(grad_norm_dict: Dict[str, float]) → None

Override this method to change the default behaviour of log_grad_norm.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

grad_norm_dict – Dictionary containing current grad norm metrics

Example:

# DEFAULT
def log_grad_norm(self, grad_norm_dict):
    self.log_dict(grad_norm_dict, on_step=True, on_epoch=True, prog_bar=False, logger=True)

property logger: Optional[Logger]

Reference to the logger object in the Trainer.

property loggers: List[Logger]

Reference to the list of loggers in the Trainer.

loss(views, **kwargs)

Loss for model :param views: :param args: :param kwargs: :return:

lr_scheduler_step(scheduler: Union[_LRScheduler, ReduceLROnPlateau], optimizer_idx: int, metric: Optional[Any]) → None

Override this method to adjust the default way the Trainer calls each scheduler. By default, Lightning calls step() and as shown in the example for each scheduler based on its interval.

Parameters

• scheduler – Learning rate scheduler.

• optimizer_idx – Index of the optimizer associated with this scheduler.

• metric – Value of the monitor used for schedulers like ReduceLROnPlateau.

Examples:

# DEFAULT
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    if metric is None:
        scheduler.step()
    else:
        scheduler.step(metric)

# Alternative way to update schedulers if it requires an epoch value
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    scheduler.step(epoch=self.current_epoch)

lr_schedulers() → Union[None, List[Union[_LRScheduler, ReduceLROnPlateau]], _LRScheduler, ReduceLROnPlateau]

Returns the learning rate scheduler(s) that are being used during training. Useful for manual optimization.

Returns

A single scheduler, or a list of schedulers in case multiple ones are present, or None if no schedulers were returned in configure_optimizers().

manual_backward(loss: Tensor, *args: Any, **kwargs: Any) → None

Call this directly from your training_step() when doing optimizations manually. By using this, Lightning can ensure that all the proper scaling gets applied when using mixed precision.

See manual optimization for more examples.

Example:

def training_step(...):
    opt = self.optimizers()
    loss = ...
    opt.zero_grad()
    # automatically applies scaling, etc...
    self.manual_backward(loss)
    opt.step()

Parameters

• loss – The tensor on which to compute gradients. Must have a graph attached.

• *args – Additional positional arguments to be forwarded to backward()

• **kwargs – Additional keyword arguments to be forwarded to backward()

modules() → Iterator[Module]

Returns an iterator over all modules in the network.

Yields

Module – a module in the network

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.modules()):
...     print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Tensor]]

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Parameters

• prefix (str) – prefix to prepend to all buffer names.

• recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

(str, torch.Tensor) – Tuple containing the name and buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, buf in self.named_buffers():
>>>    if name in ['running_var']:
>>>        print(buf.size())

named_children() → Iterator[Tuple[str, Module]]

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields

(str, Module) – Tuple containing a name and child module

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, module in model.named_children():
>>>     if name in ['conv4', 'conv5']:
>>>         print(module)

named_modules(memo: Optional[Set[Module]] = None, prefix: str = '', remove_duplicate: bool = True)

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Parameters

• memo – a memo to store the set of modules already added to the result

• prefix – a prefix that will be added to the name of the module

• remove_duplicate – whether to remove the duplicated module instances in the result or not

Yields

(str, Module) – Tuple of name and module

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.named_modules()):
...     print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Parameter]]

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Parameters

• prefix (str) – prefix to prepend to all parameter names.

• recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

(str, Parameter) – Tuple containing the name and parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, param in self.named_parameters():
>>>    if name in ['bias']:
>>>        print(param.size())

on_after_backward() → None

Called after loss.backward() and before optimizers are stepped.

Note

If using native AMP, the gradients will not be unscaled at this point. Use the on_before_optimizer_step if you need the unscaled gradients.

on_after_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch after it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_after_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = gpu_transforms(batch['x'])
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• on_before_batch_transfer()

• transfer_batch_to_device()

on_before_backward(loss: Tensor) → None

Called before loss.backward().

Parameters

loss – Loss divided by number of batches for gradient accumulation and scaled if using native AMP.

on_before_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch before it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_before_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = transforms(batch['x'])
    return batch

See also

• on_after_batch_transfer()

• transfer_batch_to_device()

on_before_optimizer_step(optimizer: Optimizer, optimizer_idx: int) → None

Called before optimizer.step().

If using gradient accumulation, the hook is called once the gradients have been accumulated. See: :paramref:`~pytorch_lightning.trainer.Trainer.accumulate_grad_batches`.

If using native AMP, the loss will be unscaled before calling this hook. See these docs for more information on the scaling of gradients.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

Example:

def on_before_optimizer_step(self, optimizer, optimizer_idx):
    # example to inspect gradient information in tensorboard
    if self.trainer.global_step % 25 == 0:  # don't make the tf file huge
        for k, v in self.named_parameters():
            self.logger.experiment.add_histogram(
                tag=k, values=v.grad, global_step=self.trainer.global_step
            )

on_before_zero_grad(optimizer: Optimizer) → None

Called after training_step() and before optimizer.zero_grad().

Called in the training loop after taking an optimizer step and before zeroing grads. Good place to inspect weight information with weights updated.

This is where it is called:

for optimizer in optimizers:
    out = training_step(...)

    model.on_before_zero_grad(optimizer) # < ---- called here
    optimizer.zero_grad()

    backward()

Parameters

optimizer – The optimizer for which grads should be zeroed.

on_fit_end() → None

Called at the very end of fit.

If on DDP it is called on every process

on_fit_start() → None

Called at the very beginning of fit.

If on DDP it is called on every process

property on_gpu: bool

Returns True if this model is currently located on a GPU.

Useful to set flags around the LightningModule for different CPU vs GPU behavior.

on_load_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning to restore your model. If you saved something with on_save_checkpoint() this is your chance to restore this.

Parameters

checkpoint – Loaded checkpoint

Example:

def on_load_checkpoint(self, checkpoint):
    # 99% of the time you don't need to implement this method
    self.something_cool_i_want_to_save = checkpoint['something_cool_i_want_to_save']

Note

Lightning auto-restores global step, epoch, and train state including amp scaling. There is no need for you to restore anything regarding training.

on_predict_batch_end(outputs: Optional[Any], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop after the batch.

Parameters

• outputs – The outputs of predict_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_end() → None

Called at the end of predicting.

on_predict_epoch_end(results: List[Any]) → None

Called at the end of predicting.

on_predict_epoch_start() → None

Called at the beginning of predicting.

on_predict_model_eval() → None

Sets the model to eval during the predict loop.

on_predict_start() → None

Called at the beginning of predicting.

on_save_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning when saving a checkpoint to give you a chance to store anything else you might want to save.

Parameters

checkpoint – The full checkpoint dictionary before it gets dumped to a file. Implementations of this hook can insert additional data into this dictionary.

Example:

def on_save_checkpoint(self, checkpoint):
    # 99% of use cases you don't need to implement this method
    checkpoint['something_cool_i_want_to_save'] = my_cool_pickable_object

Note

Lightning saves all aspects of training (epoch, global step, etc…) including amp scaling. There is no need for you to store anything about training.

on_test_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop after the batch.

Parameters

• outputs – The outputs of test_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_end() → None

Called at the end of testing.

on_test_epoch_end() → None

Called in the test loop at the very end of the epoch.

on_test_epoch_start() → None

Called in the test loop at the very beginning of the epoch.

on_test_model_eval() → None

Sets the model to eval during the test loop.

on_test_model_train() → None

Sets the model to train during the test loop.

on_test_start() → None

Called at the beginning of testing.

on_train_batch_end(outputs: Union[Tensor, Dict[str, Any]], batch: Any, batch_idx: int) → None

Called in the training loop after the batch.

Parameters

• outputs – The outputs of training_step_end(training_step(x))

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_batch_start(batch: Any, batch_idx: int) → Optional[int]

Called in the training loop before anything happens for that batch.

If you return -1 here, you will skip training for the rest of the current epoch.

Parameters

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_end() → None

Called at the end of training before logger experiment is closed.

on_train_epoch_end() → None

Called in the training loop at the very end of the epoch.

To access all batch outputs at the end of the epoch, either:

1. Implement training_epoch_end in the LightningModule OR

2. Cache data across steps on the attribute(s) of the LightningModule and access them in this hook

on_train_epoch_start() → None

Called in the training loop at the very beginning of the epoch.

on_train_start() → None

Called at the beginning of training after sanity check.

on_validation_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop after the batch.

Parameters

• outputs – The outputs of validation_step_end(validation_step(x))

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_end() → None

Called at the end of validation.

on_validation_epoch_end() → None

Called in the validation loop at the very end of the epoch.

on_validation_epoch_start() → None

Called in the validation loop at the very beginning of the epoch.

on_validation_model_eval() → None

Sets the model to eval during the val loop.

on_validation_model_train() → None

Sets the model to train during the val loop.

on_validation_start() → None

Called at the beginning of validation.

optimizer_step(epoch: int, batch_idx: int, optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int = 0, optimizer_closure: Optional[Callable[[], Any]] = None, on_tpu: bool = False, using_native_amp: bool = False, using_lbfgs: bool = False) → None

Override this method to adjust the default way the Trainer calls each optimizer.

By default, Lightning calls step() and zero_grad() as shown in the example once per optimizer. This method (and zero_grad()) won’t be called during the accumulation phase when Trainer(accumulate_grad_batches != 1). Overriding this hook has no benefit with manual optimization.

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers, this indexes into that list.

• optimizer_closure – The optimizer closure. This closure must be executed as it includes the calls to training_step(), optimizer.zero_grad(), and backward().

• on_tpu – True if TPU backward is required

• using_native_amp – True if using native amp

• using_lbfgs – True if the matching optimizer is torch.optim.LBFGS

Examples:

# DEFAULT
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    optimizer.step(closure=optimizer_closure)

# Alternating schedule for optimizer steps (i.e.: GANs)
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    # update generator opt every step
    if optimizer_idx == 0:
        optimizer.step(closure=optimizer_closure)

    # update discriminator opt every 2 steps
    if optimizer_idx == 1:
        if (batch_idx + 1) % 2 == 0 :
            optimizer.step(closure=optimizer_closure)
        else:
            # call the closure by itself to run `training_step` + `backward` without an optimizer step
            optimizer_closure()

    # ...
    # add as many optimizers as you want

Here’s another example showing how to use this for more advanced things such as learning rate warm-up:

# learning rate warm-up
def optimizer_step(
    self,
    epoch,
    batch_idx,
    optimizer,
    optimizer_idx,
    optimizer_closure,
    on_tpu,
    using_native_amp,
    using_lbfgs,
):
    # update params
    optimizer.step(closure=optimizer_closure)

    # manually warm up lr without a scheduler
    if self.trainer.global_step < 500:
        lr_scale = min(1.0, float(self.trainer.global_step + 1) / 500.0)
        for pg in optimizer.param_groups:
            pg["lr"] = lr_scale * self.learning_rate

optimizer_zero_grad(epoch: int, batch_idx: int, optimizer: Optimizer, optimizer_idx: int) → None

Override this method to change the default behaviour of optimizer.zero_grad().

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers this indexes into that list.

Examples:

# DEFAULT
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad()

# Set gradients to `None` instead of zero to improve performance.
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad(set_to_none=True)

See torch.optim.Optimizer.zero_grad() for the explanation of the above example.

optimizers(use_pl_optimizer: bool = True) → Union[Optimizer, LightningOptimizer, List[Optimizer], List[LightningOptimizer]]

Returns the optimizer(s) that are being used during training. Useful for manual optimization.

Parameters

use_pl_optimizer – If True, will wrap the optimizer(s) in a LightningOptimizer for automatic handling of precision and profiling.

Returns

A single optimizer, or a list of optimizers in case multiple ones are present.

pairwise_correlations(loader: DataLoader, train=False)

Calculates correlation for entire batch from dataloader

Parameters

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

Returns

by default returns the average pairwise correlation in each dimension (for 2 views just the correlation)

parameters(recurse: bool = True) → Iterator[Parameter]

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Parameters

recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

Parameter – module parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for param in model.parameters():
>>>     print(type(param), param.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

predict_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for prediction.

It’s recommended that all data downloads and preparation happen in prepare_data().

• predict()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying prediction samples.

Note

In the case where you return multiple prediction dataloaders, the predict_step() will have an argument dataloader_idx which matches the order here.

predict_step(batch: Any, batch_idx: int, dataloader_idx: int = 0) → Any

Step function called during predict(). By default, it calls forward(). Override to add any processing logic.

The predict_step() is used to scale inference on multi-devices.

To prevent an OOM error, it is possible to use BasePredictionWriter callback to write the predictions to disk or database after each batch or on epoch end.

The BasePredictionWriter should be used while using a spawn based accelerator. This happens for Trainer(strategy="ddp_spawn") or training on 8 TPU cores with Trainer(accelerator="tpu", devices=8) as predictions won’t be returned.

Example

class MyModel(LightningModule):

    def predict_step(self, batch, batch_idx, dataloader_idx=0):
        return self(batch)

dm = ...
model = MyModel()
trainer = Trainer(accelerator="gpu", devices=2)
predictions = trainer.predict(model, dm)

Parameters

• batch – Current batch.

• batch_idx – Index of current batch.

• dataloader_idx – Index of the current dataloader.

Returns

Predicted output

prepare_data() → None

Use this to download and prepare data. Downloading and saving data with multiple processes (distributed settings) will result in corrupted data. Lightning ensures this method is called only within a single process, so you can safely add your downloading logic within.

Warning

DO NOT set state to the model (use setup instead) since this is NOT called on every device

Example:

def prepare_data(self):
    # good
    download_data()
    tokenize()
    etc()

    # bad
    self.split = data_split
    self.some_state = some_other_state()

In a distributed environment, prepare_data can be called in two ways (using prepare_data_per_node)

1. Once per node. This is the default and is only called on LOCAL_RANK=0.

2. Once in total. Only called on GLOBAL_RANK=0.

Example:

# DEFAULT
# called once per node on LOCAL_RANK=0 of that node
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = True


# call on GLOBAL_RANK=0 (great for shared file systems)
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = False

This is called before requesting the dataloaders:

model.prepare_data()
initialize_distributed()
model.setup(stage)
model.train_dataloader()
model.val_dataloader()
model.test_dataloader()
model.predict_dataloader()

print(*args: Any, **kwargs: Any) → None

Prints only from process 0. Use this in any distributed mode to log only once.

Parameters

• *args – The thing to print. The same as for Python’s built-in print function.

• **kwargs – The same as for Python’s built-in print function.

Example:

def forward(self, x):
    self.print(x, 'in forward')

register_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

This function is deprecated in favor of register_full_backward_hook() and the behavior of this function will change in future versions.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_buffer(name: str, tensor: Optional[Tensor], persistent: bool = True) → None

Adds a buffer to the module.

This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the module’s state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module’s state_dict.

Buffers can be accessed as attributes using given names.

Parameters

• name (str) – name of the buffer. The buffer can be accessed from this module using the given name

• tensor (Tensor or None) – buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module’s state_dict.

• persistent (bool) – whether the buffer is part of this module’s state_dict.

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> self.register_buffer('running_mean', torch.zeros(num_features))

register_forward_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output. It should have the following signature:

hook(module, input, output) -> None or modified output

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward() is called.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_forward_pre_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked. It should have the following signature:

hook(module, input) -> None or modified input

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned(unless that value is already a tuple).

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_full_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

The hook will be called every time the gradients with respect to module inputs are computed. The hook should have the following signature:

hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module’s forward function.

Warning

Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_load_state_dict_post_hook(hook)

Registers a post hook to be run after module’s load_state_dict is called.

It should have the following signature::

hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict() with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearning out both missing and unexpected keys will avoid an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_module(name: str, module: Optional[Module]) → None

Alias for add_module().

register_parameter(name: str, param: Optional[Parameter]) → None

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Parameters

• name (str) – name of the parameter. The parameter can be accessed from this module using the given name

• param (Parameter or None) – parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module’s state_dict.

requires_grad_(requires_grad: bool = True) → T

Change if autograd should record operations on parameters in this module.

This method sets the parameters’ requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See Locally disabling gradient computation for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Parameters

requires_grad (bool) – whether autograd should record operations on parameters in this module. Default: True.

Returns

self

Return type

Module

save_hyperparameters(*args: Any, ignore: Optional[Union[Sequence[str], str]] = None, frame: Optional[frame] = None, logger: bool = True) → None

Save arguments to hparams attribute.

Parameters

• args – single object of dict, NameSpace or OmegaConf or string names or arguments from class __init__

• ignore – an argument name or a list of argument names from class __init__ to be ignored

• frame – a frame object. Default is None

• logger – Whether to send the hyperparameters to the logger. Default: True

Example::

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # manually assign arguments
...         self.save_hyperparameters('arg1', 'arg3')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

>>> class AutomaticArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # equivalent automatic
...         self.save_hyperparameters()
...     def forward(self, *args, **kwargs):
...         ...
>>> model = AutomaticArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg2": abc
"arg3": 3.14

>>> class SingleArgModel(HyperparametersMixin):
...     def __init__(self, params):
...         super().__init__()
...         # manually assign single argument
...         self.save_hyperparameters(params)
...     def forward(self, *args, **kwargs):
...         ...
>>> model = SingleArgModel(Namespace(p1=1, p2='abc', p3=3.14))
>>> model.hparams
"p1": 1
"p2": abc
"p3": 3.14

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # pass argument(s) to ignore as a string or in a list
...         self.save_hyperparameters(ignore='arg2')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

score(loader: DataLoader, **kwargs)

Returns average correlation in each dimension (averages over all pairs for multiview)

Parameters

• **kwargs –

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

set_extra_state(state: Any)

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Parameters

state (dict) – Extra state from the state_dict

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters

**params (dict) – Estimator parameters.

Returns

self – Estimator instance.

Return type

estimator instance

setup(stage: str) → None

Called at the beginning of fit (train + validate), validate, test, or predict. This is a good hook when you need to build models dynamically or adjust something about them. This hook is called on every process when using DDP.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

Example:

class LitModel(...):
    def __init__(self):
        self.l1 = None

    def prepare_data(self):
        download_data()
        tokenize()

        # don't do this
        self.something = else

    def setup(self, stage):
        data = load_data(...)
        self.l1 = nn.Linear(28, data.num_classes)

share_memory() → T

See torch.Tensor.share_memory_()

state_dict(*args, destination=None, prefix='', keep_vars=False)

Returns a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

Note

The returned object is a shallow copy. It contains references to the module’s parameters and buffers.

Warning

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

Warning

Please avoid the use of argument destination as it is not designed for end-users.

Parameters

• destination (dict, optional) – If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned. Default: None.

• prefix (str, optional) – a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.

• keep_vars (bool, optional) – by default the Tensor s returned in the state dict are detached from autograd. If it’s set to True, detaching will not be performed. Default: False.

Returns

a dictionary containing a whole state of the module

Return type

dict

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> module.state_dict().keys()
['bias', 'weight']

tbptt_split_batch(batch: Any, split_size: int) → List[Any]

When using truncated backpropagation through time, each batch must be split along the time dimension. Lightning handles this by default, but for custom behavior override this function.

Parameters

• batch – Current batch

• split_size – The size of the split

Returns

List of batch splits. Each split will be passed to training_step() to enable truncated back propagation through time. The default implementation splits root level Tensors and Sequences at dim=1 (i.e. time dim). It assumes that each time dim is the same length.

Examples:

def tbptt_split_batch(self, batch, split_size):
    splits = []
    for t in range(0, time_dims[0], split_size):
        batch_split = []
        for i, x in enumerate(batch):
            if isinstance(x, torch.Tensor):
                split_x = x[:, t:t + split_size]
            elif isinstance(x, collections.abc.Sequence):
                split_x = [None] * len(x)
                for batch_idx in range(len(x)):
                  split_x[batch_idx] = x[batch_idx][t:t + split_size]
            batch_split.append(split_x)
        splits.append(batch_split)
    return splits

Note

Called in the training loop after on_train_batch_start() if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0. Each returned batch split is passed separately to training_step().

teardown(stage: str) → None

Called at the end of fit (train + validate), validate, test, or predict.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

test_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for testing.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• test()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying testing samples.

Example:

def test_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def test_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a test dataset and a test_step(), you don’t need to implement this method.

Note

In the case where you return multiple test dataloaders, the test_step() will have an argument dataloader_idx which matches the order here.

test_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of a test epoch with the output of all test steps.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

outputs – List of outputs you defined in test_step_end(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader

Returns

None

Note

If you didn’t define a test_step(), this won’t be called.

Examples

With a single dataloader:

def test_epoch_end(self, outputs):
    # do something with the outputs of all test batches
    all_test_preds = test_step_outputs.predictions

    some_result = calc_all_results(all_test_preds)
    self.log(some_result)

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each test step for that dataloader.

def test_epoch_end(self, outputs):
    final_value = 0
    for dataloader_outputs in outputs:
        for test_step_out in dataloader_outputs:
            # do something
            final_value += test_step_out

    self.log("final_metric", final_value)

test_step(batch, batch_idx)

Operates on a single batch of data from the test set. In this step you’d normally generate examples or calculate anything of interest such as accuracy.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_id – The index of the dataloader that produced this batch. (only if multiple test dataloaders used).

Returns

Any of.

• Any object or value

• None - Testing will skip to the next batch

# if you have one test dataloader:
def test_step(self, batch, batch_idx):
    ...


# if you have multiple test dataloaders:
def test_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single test dataset
def test_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    test_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'test_loss': loss, 'test_acc': test_acc})

If you pass in multiple test dataloaders, test_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple test dataloaders
def test_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to test you don’t need to implement this method.

Note

When the test_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of the test epoch, the model goes back to training mode and gradients are enabled.

test_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when testing with DP because test_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [test_step(sub_batch) for sub_batch in sub_batches]
test_step_end(step_output)

Parameters

step_output – What you return in test_step() for each batch part.

Returns

None or anything

# WITHOUT test_step_end
# if used in DP, this batch is 1/num_gpus large
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    loss = self.softmax(out)
    self.log("test_loss", loss)


# --------------
# with test_step_end to do softmax over the full batch
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return out


def test_step_end(self, output_results):
    # this out is now the full size of the batch
    all_test_step_outs = output_results.out
    loss = nce_loss(all_test_step_outs)
    self.log("test_loss", loss)

See also

See the Multi GPU Training guide for more details.

to(*args: Any, **kwargs: Any) → Self

See torch.nn.Module.to().

to_empty(*, device: Union[str, device]) → T

Moves the parameters and buffers to the specified device without copying storage.

Parameters

device (torch.device) – The desired device of the parameters and buffers in this module.

Returns

self

Return type

Module

to_onnx(file_path: Union[str, Path], input_sample: Optional[Any] = None, **kwargs: Any) → None

Saves the model in ONNX format.

Parameters

• file_path – The path of the file the onnx model should be saved to.

• input_sample – An input for tracing. Default: None (Use self.example_input_array)

• **kwargs – Will be passed to torch.onnx.export function.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))

>>> with tempfile.NamedTemporaryFile(suffix='.onnx', delete=False) as tmpfile:
...     model = SimpleModel()
...     input_sample = torch.randn((1, 64))
...     model.to_onnx(tmpfile.name, input_sample, export_params=True)
...     os.path.isfile(tmpfile.name)
True

to_torchscript(file_path: Optional[Union[str, Path]] = None, method: Optional[str] = 'script', example_inputs: Optional[Any] = None, **kwargs: Any) → Union[ScriptModule, Dict[str, ScriptModule]]

By default compiles the whole model to a ScriptModule. If you want to use tracing, please provided the argument method='trace' and make sure that either the example_inputs argument is provided, or the model has example_input_array set. If you would like to customize the modules that are scripted you should override this method. In case you want to return multiple modules, we recommend using a dictionary.

Parameters

• file_path – Path where to save the torchscript. Default: None (no file saved).

• method – Whether to use TorchScript’s script or trace method. Default: ‘script’

• example_inputs – An input to be used to do tracing when method is set to ‘trace’. Default: None (uses example_input_array)

• **kwargs – Additional arguments that will be passed to the torch.jit.script() or torch.jit.trace() function.

Note

• Requires the implementation of the forward() method.

• The exported script will be set to evaluation mode.

• It is recommended that you install the latest supported version of PyTorch to use this feature without limitations. See also the torch.jit documentation for supported features.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))
...
>>> model = SimpleModel()
>>> model.to_torchscript(file_path="model.pt")  
>>> os.path.isfile("model.pt")  
>>> torch.jit.save(model.to_torchscript(file_path="model_trace.pt", method='trace', 
...                                     example_inputs=torch.randn(1, 64)))  
>>> os.path.isfile("model_trace.pt")  
True

Returns

This LightningModule as a torchscript, regardless of whether file_path is defined or not.

toggle_optimizer(optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int) → None

Makes sure only the gradients of the current optimizer’s parameters are calculated in the training step to prevent dangling gradients in multiple-optimizer setup.

This is only called automatically when automatic optimization is enabled and multiple optimizers are used. It works with untoggle_optimizer() to make sure param_requires_grad_state is properly reset.

Parameters

• optimizer – The optimizer to toggle.

• optimizer_idx – The index of the optimizer to toggle.

train(mode: bool = True) → T

Sets the module in training mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

Parameters

mode (bool) – whether to set training mode (True) or evaluation mode (False). Default: True.

Returns

self

Return type

Module

train_dataloader() → Union[DataLoader, Sequence[DataLoader], Sequence[Sequence[DataLoader]], Sequence[Dict[str, DataLoader]], Dict[str, DataLoader], Dict[str, Dict[str, DataLoader]], Dict[str, Sequence[DataLoader]]]

Implement one or more PyTorch DataLoaders for training.

Returns

A collection of torch.utils.data.DataLoader specifying training samples. In the case of multiple dataloaders, please see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• fit()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Example:

# single dataloader
def train_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=True, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=True
    )
    return loader

# multiple dataloaders, return as list
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a list of tensors: [batch_mnist, batch_cifar]
    return [mnist_loader, cifar_loader]

# multiple dataloader, return as dict
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a dict of tensors: {'mnist': batch_mnist, 'cifar': batch_cifar}
    return {'mnist': mnist_loader, 'cifar': cifar_loader}

training_epoch_end(outputs: List[Union[Tensor, Dict[str, Any]]]) → None

Called at the end of the training epoch with the outputs of all training steps. Use this in case you need to do something with all the outputs returned by training_step().

# the pseudocode for these calls
train_outs = []
for train_batch in train_data:
    out = training_step(train_batch)
    train_outs.append(out)
training_epoch_end(train_outs)

Parameters

outputs – List of outputs you defined in training_step(). If there are multiple optimizers or when using truncated_bptt_steps > 0, the lists have the dimensions (n_batches, tbptt_steps, n_optimizers). Dimensions of length 1 are squeezed.

Returns

None

Note

If this method is not overridden, this won’t be called.

def training_epoch_end(self, training_step_outputs):
    # do something with all training_step outputs
    for out in training_step_outputs:
        ...

training_step(batch, batch_idx)

Here you compute and return the training loss and some additional metrics for e.g. the progress bar or logger.

Parameters

• batch (Tensor | (Tensor, …) | [Tensor, …]) – The output of your DataLoader. A tensor, tuple or list.

• batch_idx (int) – Integer displaying index of this batch

• optimizer_idx (int) – When using multiple optimizers, this argument will also be present.

• hiddens (Any) – Passed in if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0.

Returns

Any of.

• Tensor - The loss tensor

• dict - A dictionary. Can include any keys, but must include the key 'loss'

• None - Training will skip to the next batch. This is only for automatic optimization.

  This is not supported for multi-GPU, TPU, IPU, or DeepSpeed.

In this step you’d normally do the forward pass and calculate the loss for a batch. You can also do fancier things like multiple forward passes or something model specific.

Example:

def training_step(self, batch, batch_idx):
    x, y, z = batch
    out = self.encoder(x)
    loss = self.loss(out, x)
    return loss

If you define multiple optimizers, this step will be called with an additional optimizer_idx parameter.

# Multiple optimizers (e.g.: GANs)
def training_step(self, batch, batch_idx, optimizer_idx):
    if optimizer_idx == 0:
        # do training_step with encoder
        ...
    if optimizer_idx == 1:
        # do training_step with decoder
        ...

If you add truncated back propagation through time you will also get an additional argument with the hidden states of the previous step.

# Truncated back-propagation through time
def training_step(self, batch, batch_idx, hiddens):
    # hiddens are the hidden states from the previous truncated backprop step
    out, hiddens = self.lstm(data, hiddens)
    loss = ...
    return {"loss": loss, "hiddens": hiddens}

Note

The loss value shown in the progress bar is smoothed (averaged) over the last values, so it differs from the actual loss returned in train/validation step.

Note

When accumulate_grad_batches > 1, the loss returned here will be automatically normalized by accumulate_grad_batches internally.

training_step_end(step_output: Union[Tensor, Dict[str, Any]]) → Union[Tensor, Dict[str, Any]]

Use this when training with dp because training_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [training_step(sub_batch) for sub_batch in sub_batches]
training_step_end(step_output)

Parameters

step_output – What you return in training_step for each batch part.

Returns

Anything

When using the DP strategy, only a portion of the batch is inside the training_step:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)

    # softmax uses only a portion of the batch in the denominator
    loss = self.softmax(out)
    loss = nce_loss(loss)
    return loss

If you wish to do something with all the parts of the batch, then use this method to do it:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return {"pred": out}


def training_step_end(self, training_step_outputs):
    gpu_0_pred = training_step_outputs[0]["pred"]
    gpu_1_pred = training_step_outputs[1]["pred"]
    gpu_n_pred = training_step_outputs[n]["pred"]

    # this softmax now uses the full batch
    loss = nce_loss([gpu_0_pred, gpu_1_pred, gpu_n_pred])
    return loss

See also

See the Multi GPU Training guide for more details.

transfer_batch_to_device(batch: Any, device: device, dataloader_idx: int) → Any

Override this hook if your DataLoader returns tensors wrapped in a custom data structure.

The data types listed below (and any arbitrary nesting of them) are supported out of the box:

• torch.Tensor or anything that implements .to(…)

• list

• dict

• tuple

For anything else, you need to define how the data is moved to the target device (CPU, GPU, TPU, …).

Note

This hook should only transfer the data and not modify it, nor should it move the data to any other device than the one passed in as argument (unless you know what you are doing). To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be transferred to a new device.

• device – The target device as defined in PyTorch.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A reference to the data on the new device.

Example:

def transfer_batch_to_device(self, batch, device, dataloader_idx):
    if isinstance(batch, CustomBatch):
        # move all tensors in your custom data structure to the device
        batch.samples = batch.samples.to(device)
        batch.targets = batch.targets.to(device)
    elif dataloader_idx == 0:
        # skip device transfer for the first dataloader or anything you wish
        pass
    else:
        batch = super().transfer_batch_to_device(data, device, dataloader_idx)
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• move_data_to_device()

• apply_to_collection()

transform(loader: DataLoader)

Parameters

loader – a dataloader that matches the structure of that used for training

Returns

transformed views

property truncated_bptt_steps: int

Enables Truncated Backpropagation Through Time in the Trainer when set to a positive integer.

It represents the number of times training_step() gets called before backpropagation. If this is > 0, the training_step() receives an additional argument hiddens and is expected to return a hidden state.

type(dst_type: Union[str, dtype]) → Self

See torch.nn.Module.type().

unfreeze() → None

Unfreeze all parameters for training.

model = MyLightningModule(...)
model.unfreeze()

untoggle_optimizer(optimizer_idx: int) → None

Resets the state of required gradients that were toggled with toggle_optimizer().

This is only called automatically when automatic optimization is enabled and multiple optimizers are used.

Parameters

optimizer_idx – The index of the optimizer to untoggle.

val_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for validation.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

It’s recommended that all data downloads and preparation happen in prepare_data().

• fit()

• validate()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying validation samples.

Examples:

def val_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False,
                    transform=transform, download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def val_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a validation dataset and a validation_step(), you don’t need to implement this method.

Note

In the case where you return multiple validation dataloaders, the validation_step() will have an argument dataloader_idx which matches the order here.

validation_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of the validation epoch with the outputs of all validation steps.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

outputs – List of outputs you defined in validation_step(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader.

Returns

None

Note

If you didn’t define a validation_step(), this won’t be called.

Examples

With a single dataloader:

def validation_epoch_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each validation step for that dataloader.

def validation_epoch_end(self, outputs):
    for dataloader_output_result in outputs:
        dataloader_outs = dataloader_output_result.dataloader_i_outputs

    self.log("final_metric", final_value)

validation_step(batch, batch_idx)

Operates on a single batch of data from the validation set. In this step you’d might generate examples or calculate anything of interest like accuracy.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_idx – The index of the dataloader that produced this batch. (only if multiple val dataloaders used)

Returns

• Any object or value

• None - Validation will skip to the next batch

# pseudocode of order
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    if defined("validation_step_end"):
        out = validation_step_end(out)
    val_outs.append(out)
val_outs = validation_epoch_end(val_outs)

# if you have one val dataloader:
def validation_step(self, batch, batch_idx):
    ...


# if you have multiple val dataloaders:
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single validation dataset
def validation_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    val_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'val_loss': loss, 'val_acc': val_acc})

If you pass in multiple val dataloaders, validation_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple validation dataloaders
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to validate you don’t need to implement this method.

Note

When the validation_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of validation, the model goes back to training mode and gradients are enabled.

validation_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when validating with dp because validation_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [validation_step(sub_batch) for sub_batch in sub_batches]
validation_step_end(step_output)

Parameters

step_output – What you return in validation_step() for each batch part.

Returns

None or anything

# WITHOUT validation_step_end
# if used in DP, this batch is 1/num_gpus large
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    loss = self.softmax(out)
    loss = nce_loss(loss)
    self.log("val_loss", loss)


# --------------
# with validation_step_end to do softmax over the full batch
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    return out


def validation_step_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

See also

See the Multi GPU Training guide for more details.

xpu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the XPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

zero_grad(set_to_none: bool = False) → None

Sets gradients of all model parameters to zero. See similar function under torch.optim.Optimizer for more context.

Parameters

set_to_none (bool) – instead of setting to zero, set the grads to None. See torch.optim.Optimizer.zero_grad() for details.

class cca_zoo.deepmodels.DCCA_EigenGame(latent_dims: int, encoders=None, r: float = 0, **kwargs)

Bases: DCCA

References

Chapman, James, Ana Lawry Aguila, and Lennie Wells. “A Generalized EigenGame with Extensions to Multiview Representation Learning.” arXiv preprint arXiv:2211.11323 (2022).

forward(views, **kwargs)

We use the forward model to define the transformation of views to the latent space :param views: batches for each view separated by commas

loss(views, **kwargs)

Loss for model :param views: :param args: :param kwargs: :return:

add_module(name: str, module: Optional[Module]) → None

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Parameters

• name (str) – name of the child module. The child module can be accessed from this module using the given name

• module (Module) – child module to be added to the module.

all_gather(data: Union[Tensor, Dict, List, Tuple], group: Optional[Any] = None, sync_grads: bool = False) → Union[Tensor, Dict, List, Tuple]

Allows users to call self.all_gather() from the LightningModule, thus making the all_gather operation accelerator agnostic. all_gather is a function provided by accelerators to gather a tensor from several distributed processes.

Parameters

• data – int, float, tensor of shape (batch, …), or a (possibly nested) collection thereof.

• group – the process group to gather results from. Defaults to all processes (world)

• sync_grads – flag that allows users to synchronize gradients for the all_gather operation

Returns

A tensor of shape (world_size, batch, …), or if the input was a collection the output will also be a collection with tensors of this shape.

apply(fn: Callable[[Module], None]) → T

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also torch.nn.init).

Parameters

fn (Module -> None) – function to be applied to each submodule

Returns

self

Return type

Module

Example:

>>> @torch.no_grad()
>>> def init_weights(m):
>>>     print(m)
>>>     if type(m) == nn.Linear:
>>>         m.weight.fill_(1.0)
>>>         print(m.weight)
>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

property automatic_optimization: bool

If set to False you are responsible for calling .backward(), .step(), .zero_grad().

backward(loss: Tensor, optimizer: Optional[Steppable], optimizer_idx: Optional[int], *args: Any, **kwargs: Any) → None

Called to perform backward on the loss returned in training_step(). Override this hook with your own implementation if you need to.

Parameters

• loss – The loss tensor returned by training_step(). If gradient accumulation is used, the loss here holds the normalized value (scaled by 1 / accumulation steps).

• optimizer – Current optimizer being used. None if using manual optimization.

• optimizer_idx – Index of the current optimizer being used. None if using manual optimization.

Example:

def backward(self, loss, optimizer, optimizer_idx):
    loss.backward()

bfloat16() → T

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns

self

Return type

Module

buffers(recurse: bool = True) → Iterator[Tensor]

Returns an iterator over module buffers.

Parameters

recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

torch.Tensor – module buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for buf in model.buffers():
>>>     print(type(buf), buf.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

children() → Iterator[Module]

Returns an iterator over immediate children modules.

Yields

Module – a child module

clip_gradients(optimizer: Optimizer, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Handles gradient clipping internally.

Note

Do not override this method. If you want to customize gradient clipping, consider using configure_gradient_clipping() method.

Parameters

• optimizer – Current optimizer being used.

• gradient_clip_val – The value at which to clip gradients.

• gradient_clip_algorithm – The gradient clipping algorithm to use. Pass gradient_clip_algorithm="value" to clip by value, and gradient_clip_algorithm="norm" to clip by norm.

configure_callbacks()

Configure model-specific callbacks. When the model gets attached, e.g., when .fit() or .test() gets called, the list or a callback returned here will be merged with the list of callbacks passed to the Trainer’s callbacks argument. If a callback returned here has the same type as one or several callbacks already present in the Trainer’s callbacks list, it will take priority and replace them. In addition, Lightning will make sure ModelCheckpoint callbacks run last.

Returns

A callback or a list of callbacks which will extend the list of callbacks in the Trainer.

Example:

def configure_callbacks(self):
    early_stop = EarlyStopping(monitor="val_acc", mode="max")
    checkpoint = ModelCheckpoint(monitor="val_loss")
    return [early_stop, checkpoint]

configure_gradient_clipping(optimizer: Optimizer, optimizer_idx: int, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Perform gradient clipping for the optimizer parameters. Called before optimizer_step().

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

• gradient_clip_val – The value at which to clip gradients. By default value passed in Trainer will be available here.

• gradient_clip_algorithm – The gradient clipping algorithm to use. By default value passed in Trainer will be available here.

Example:

# Perform gradient clipping on gradients associated with discriminator (optimizer_idx=1) in GAN
def configure_gradient_clipping(self, optimizer, optimizer_idx, gradient_clip_val, gradient_clip_algorithm):
    if optimizer_idx == 1:
        # Lightning will handle the gradient clipping
        self.clip_gradients(
            optimizer,
            gradient_clip_val=gradient_clip_val,
            gradient_clip_algorithm=gradient_clip_algorithm
        )
    else:
        # implement your own custom logic to clip gradients for generator (optimizer_idx=0)

configure_optimizers()

Collects learnable parameters and configures the optimizer and learning rate scheduler. :returns: two lists containing the optimizer and the scheduler. :rtype: Tuple[List, List]

configure_sharded_model() → None

Hook to create modules in a distributed aware context. This is useful for when using sharded plugins, where we’d like to shard the model instantly, which is useful for extremely large models which can save memory and initialization time.

This hook is called during each of fit/val/test/predict stages in the same process, so ensure that implementation of this hook is idempotent.

cpu() → Self

See torch.nn.Module.cpu().

cuda(device: Optional[Union[device, int]] = None) → Self

Moves all model parameters and buffers to the GPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized.

Parameters

device – If specified, all parameters will be copied to that device. If None, the current CUDA device index will be used.

Returns

self

Return type

Module

property current_epoch: int

The current epoch in the Trainer, or 0 if not attached.

double() → Self

See torch.nn.Module.double().

eval() → T

Sets the module in evaluation mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False).

See Locally disabling gradient computation for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns

self

Return type

Module

property example_input_array: Optional[Union[Tensor, Tuple, Dict]]

The example input array is a specification of what the module can consume in the forward() method. The return type is interpreted as follows:

• Single tensor: It is assumed the model takes a single argument, i.e., model.forward(model.example_input_array)

• Tuple: The input array should be interpreted as a sequence of positional arguments, i.e., model.forward(*model.example_input_array)

• Dict: The input array represents named keyword arguments, i.e., model.forward(**model.example_input_array)

extra_repr() → str

Set the extra representation of the module

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

abstract fit(views: Iterable[ndarray], y=None, **kwargs)

Fits the model to the given data

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• y (None) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

self

Return type

object

fit_transform(views: Iterable[ndarray], **kwargs)

Fits the model to the given data and returns the transformed views

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• kwargs (any additional keyword arguments required by the given model) –

Returns

transformed_views

Return type

list of numpy arrays

float() → Self

See torch.nn.Module.float().

freeze() → None

Freeze all params for inference.

Example:

model = MyLightningModule(...)
model.freeze()

get_buffer(target: str) → Tensor

Returns the buffer given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The buffer referenced by target

Return type

torch.Tensor

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state() → Any

Returns any extra state to include in the module’s state_dict. Implement this and a corresponding set_extra_state() for your module if you need to store extra state. This function is called when building the module’s state_dict().

Note that extra state should be pickleable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns

Any extra state to store in the module’s state_dict

Return type

object

get_factor_loadings(views: Iterable[ndarray], normalize=True, **kwargs)

Returns the factor loadings for each view

Parameters

• views (list/tuple of numpy arrays or array likes with the same number of rows (samples)) –

• normalize (bool, optional) – Whether to normalize the factor loadings. Default is True.

• kwargs (any additional keyword arguments required by the given model) –

Returns

factor_loadings

Return type

list of numpy arrays

get_parameter(target: str) → Parameter

Returns the parameter given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The Parameter referenced by target

Return type

torch.nn.Parameter

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_params(deep=True)

Get parameters for this estimator.

Parameters

deep (bool, default=True) – If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns

params – Parameter names mapped to their values.

Return type

dict

get_submodule(target: str) → Module

Returns the submodule given by target if it exists, otherwise throws an error.

For example, let’s say you have an nn.Module A that looks like this:

A(
    (net_b): Module(
        (net_c): Module(
            (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2))
        )
        (linear): Linear(in_features=100, out_features=200, bias=True)
    )
)

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Parameters

target – The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns

The submodule referenced by target

Return type

torch.nn.Module

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Module

property global_rank: int

The index of the current process across all nodes and devices.

property global_step: int

Total training batches seen across all epochs.

If no Trainer is attached, this propery is 0.

half() → Self

See torch.nn.Module.half().

property hparams: Union[AttributeDict, MutableMapping]

The collection of hyperparameters saved with save_hyperparameters(). It is mutable by the user. For the frozen set of initial hyperparameters, use hparams_initial.

Returns

Mutable hyperparameters dictionary

property hparams_initial: AttributeDict

The collection of hyperparameters saved with save_hyperparameters(). These contents are read-only. Manual updates to the saved hyperparameters can instead be performed through hparams.

Returns

immutable initial hyperparameters

Return type

AttributeDict

ipu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the IPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on IPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

classmethod load_from_checkpoint(checkpoint_path: Union[str, Path, IO], map_location: Optional[Union[device, str, int, Callable[[Union[device, str, int]], Union[device, str, int]], Dict[Union[device, str, int], Union[device, str, int]]]] = None, hparams_file: Optional[Union[str, Path]] = None, strict: bool = True, **kwargs: Any) → Self

Primary way of loading a model from a checkpoint. When Lightning saves a checkpoint it stores the arguments passed to __init__ in the checkpoint under "hyper_parameters".

Any arguments specified through **kwargs will override args stored in "hyper_parameters".

Parameters

• checkpoint_path – Path to checkpoint. This can also be a URL, or file-like object

• map_location – If your checkpoint saved a GPU model and you now load on CPUs or a different number of GPUs, use this to map to the new setup. The behaviour is the same as in torch.load().

• hparams_file –

  Optional path to a .yaml or .csv file with hierarchical structure as in this example:

  drop_prob: 0.2
  dataloader:
      batch_size: 32
  

  You most likely won’t need this since Lightning will always save the hyperparameters to the checkpoint. However, if your checkpoint weights don’t have the hyperparameters saved, use this method to pass in a .yaml file with the hparams you’d like to use. These will be converted into a dict and passed into your LightningModule for use.

  If your model’s hparams argument is Namespace and .yaml file has hierarchical structure, you need to refactor your model to treat hparams as dict.

• strict – Whether to strictly enforce that the keys in checkpoint_path match the keys returned by this module’s state dict.

• **kwargs – Any extra keyword args needed to init the model. Can also be used to override saved hyperparameter values.

Returns

LightningModule instance with loaded weights and hyperparameters (if available).

Note

load_from_checkpoint is a class method. You should use your LightningModule class to call it instead of the LightningModule instance.

Example:

# load weights without mapping ...
model = MyLightningModule.load_from_checkpoint('path/to/checkpoint.ckpt')

# or load weights mapping all weights from GPU 1 to GPU 0 ...
map_location = {'cuda:1':'cuda:0'}
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    map_location=map_location
)

# or load weights and hyperparameters from separate files.
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    hparams_file='/path/to/hparams_file.yaml'
)

# override some of the params with new values
model = MyLightningModule.load_from_checkpoint(
    PATH,
    num_layers=128,
    pretrained_ckpt_path=NEW_PATH,
)

# predict
pretrained_model.eval()
pretrained_model.freeze()
y_hat = pretrained_model(x)

load_state_dict(state_dict: Mapping[str, Any], strict: bool = True)

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Parameters

• state_dict (dict) – a dict containing parameters and persistent buffers.

• strict (bool, optional) – whether to strictly enforce that the keys in state_dict match the keys returned by this module’s state_dict() function. Default: True

Returns

• missing_keys is a list of str containing the missing keys

• unexpected_keys is a list of str containing the unexpected keys

Return type

NamedTuple with missing_keys and unexpected_keys fields

Note

If a parameter or buffer is registered as None and its corresponding key exists in state_dict, load_state_dict() will raise a RuntimeError.

property local_rank: int

The index of the current process within a single node.

log(name: str, value: Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, metric_attribute: Optional[str] = None, rank_zero_only: bool = False) → None

Log a key, value pair.

Example:

self.log('train_loss', loss)

The default behavior per hook is documented here: extensions/logging:Automatic Logging.

Parameters

• name – key to log.

• value – value to log. Can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress bar.

• logger – if True logs to the logger.

• on_step – if True logs at this step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto detach the graph.

• sync_dist – if True, reduces the metric across devices. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the DDP group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple dataloaders). If False, user needs to give unique names for each dataloader to not mix the values.

• batch_size – Current batch_size. This will be directly inferred from the loaded batch, but for some data structures you might need to explicitly provide it.

• metric_attribute – To restore the metric state, Lightning requires the reference of the torchmetrics.Metric in your model. This is found automatically if it is a model attribute.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_dict(dictionary: Mapping[str, Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, rank_zero_only: bool = False) → None

Log a dictionary of values at once.

Example:

values = {'loss': loss, 'acc': acc, ..., 'metric_n': metric_n}
self.log_dict(values)

Parameters

• dictionary – key value pairs. The values can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress base.

• logger – if True logs to the logger.

• on_step – if True logs at this step. None auto-logs for training_step but not validation/test_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. None auto-logs for val/test step but not training_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto-detach the graph

• sync_dist – if True, reduces the metric across GPUs/TPUs. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the ddp group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple). If False, user needs to give unique names for each dataloader to not mix values.

• batch_size – Current batch size. This will be directly inferred from the loaded batch, but some data structures might need to explicitly provide it.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_grad_norm(grad_norm_dict: Dict[str, float]) → None

Override this method to change the default behaviour of log_grad_norm.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

grad_norm_dict – Dictionary containing current grad norm metrics

Example:

# DEFAULT
def log_grad_norm(self, grad_norm_dict):
    self.log_dict(grad_norm_dict, on_step=True, on_epoch=True, prog_bar=False, logger=True)

property logger: Optional[Logger]

Reference to the logger object in the Trainer.

property loggers: List[Logger]

Reference to the list of loggers in the Trainer.

lr_scheduler_step(scheduler: Union[_LRScheduler, ReduceLROnPlateau], optimizer_idx: int, metric: Optional[Any]) → None

Override this method to adjust the default way the Trainer calls each scheduler. By default, Lightning calls step() and as shown in the example for each scheduler based on its interval.

Parameters

• scheduler – Learning rate scheduler.

• optimizer_idx – Index of the optimizer associated with this scheduler.

• metric – Value of the monitor used for schedulers like ReduceLROnPlateau.

Examples:

# DEFAULT
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    if metric is None:
        scheduler.step()
    else:
        scheduler.step(metric)

# Alternative way to update schedulers if it requires an epoch value
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    scheduler.step(epoch=self.current_epoch)

lr_schedulers() → Union[None, List[Union[_LRScheduler, ReduceLROnPlateau]], _LRScheduler, ReduceLROnPlateau]

Returns the learning rate scheduler(s) that are being used during training. Useful for manual optimization.

Returns

A single scheduler, or a list of schedulers in case multiple ones are present, or None if no schedulers were returned in configure_optimizers().

manual_backward(loss: Tensor, *args: Any, **kwargs: Any) → None

Call this directly from your training_step() when doing optimizations manually. By using this, Lightning can ensure that all the proper scaling gets applied when using mixed precision.

See manual optimization for more examples.

Example:

def training_step(...):
    opt = self.optimizers()
    loss = ...
    opt.zero_grad()
    # automatically applies scaling, etc...
    self.manual_backward(loss)
    opt.step()

Parameters

• loss – The tensor on which to compute gradients. Must have a graph attached.

• *args – Additional positional arguments to be forwarded to backward()

• **kwargs – Additional keyword arguments to be forwarded to backward()

modules() → Iterator[Module]

Returns an iterator over all modules in the network.

Yields

Module – a module in the network

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.modules()):
...     print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Tensor]]

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Parameters

• prefix (str) – prefix to prepend to all buffer names.

• recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

(str, torch.Tensor) – Tuple containing the name and buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, buf in self.named_buffers():
>>>    if name in ['running_var']:
>>>        print(buf.size())

named_children() → Iterator[Tuple[str, Module]]

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields

(str, Module) – Tuple containing a name and child module

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, module in model.named_children():
>>>     if name in ['conv4', 'conv5']:
>>>         print(module)

named_modules(memo: Optional[Set[Module]] = None, prefix: str = '', remove_duplicate: bool = True)

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Parameters

• memo – a memo to store the set of modules already added to the result

• prefix – a prefix that will be added to the name of the module

• remove_duplicate – whether to remove the duplicated module instances in the result or not

Yields

(str, Module) – Tuple of name and module

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.named_modules()):
...     print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Parameter]]

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Parameters

• prefix (str) – prefix to prepend to all parameter names.

• recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

(str, Parameter) – Tuple containing the name and parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, param in self.named_parameters():
>>>    if name in ['bias']:
>>>        print(param.size())

on_after_backward() → None

Called after loss.backward() and before optimizers are stepped.

Note

If using native AMP, the gradients will not be unscaled at this point. Use the on_before_optimizer_step if you need the unscaled gradients.

on_after_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch after it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_after_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = gpu_transforms(batch['x'])
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• on_before_batch_transfer()

• transfer_batch_to_device()

on_before_backward(loss: Tensor) → None

Called before loss.backward().

Parameters

loss – Loss divided by number of batches for gradient accumulation and scaled if using native AMP.

on_before_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch before it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_before_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = transforms(batch['x'])
    return batch

See also

• on_after_batch_transfer()

• transfer_batch_to_device()

on_before_optimizer_step(optimizer: Optimizer, optimizer_idx: int) → None

Called before optimizer.step().

If using gradient accumulation, the hook is called once the gradients have been accumulated. See: :paramref:`~pytorch_lightning.trainer.Trainer.accumulate_grad_batches`.

If using native AMP, the loss will be unscaled before calling this hook. See these docs for more information on the scaling of gradients.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

Example:

def on_before_optimizer_step(self, optimizer, optimizer_idx):
    # example to inspect gradient information in tensorboard
    if self.trainer.global_step % 25 == 0:  # don't make the tf file huge
        for k, v in self.named_parameters():
            self.logger.experiment.add_histogram(
                tag=k, values=v.grad, global_step=self.trainer.global_step
            )

on_before_zero_grad(optimizer: Optimizer) → None

Called after training_step() and before optimizer.zero_grad().

Called in the training loop after taking an optimizer step and before zeroing grads. Good place to inspect weight information with weights updated.

This is where it is called:

for optimizer in optimizers:
    out = training_step(...)

    model.on_before_zero_grad(optimizer) # < ---- called here
    optimizer.zero_grad()

    backward()

Parameters

optimizer – The optimizer for which grads should be zeroed.

on_fit_end() → None

Called at the very end of fit.

If on DDP it is called on every process

on_fit_start() → None

Called at the very beginning of fit.

If on DDP it is called on every process

property on_gpu: bool

Returns True if this model is currently located on a GPU.

Useful to set flags around the LightningModule for different CPU vs GPU behavior.

on_load_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning to restore your model. If you saved something with on_save_checkpoint() this is your chance to restore this.

Parameters

checkpoint – Loaded checkpoint

Example:

def on_load_checkpoint(self, checkpoint):
    # 99% of the time you don't need to implement this method
    self.something_cool_i_want_to_save = checkpoint['something_cool_i_want_to_save']

Note

Lightning auto-restores global step, epoch, and train state including amp scaling. There is no need for you to restore anything regarding training.

on_predict_batch_end(outputs: Optional[Any], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop after the batch.

Parameters

• outputs – The outputs of predict_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_end() → None

Called at the end of predicting.

on_predict_epoch_end(results: List[Any]) → None

Called at the end of predicting.

on_predict_epoch_start() → None

Called at the beginning of predicting.

on_predict_model_eval() → None

Sets the model to eval during the predict loop.

on_predict_start() → None

Called at the beginning of predicting.

on_save_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning when saving a checkpoint to give you a chance to store anything else you might want to save.

Parameters

checkpoint – The full checkpoint dictionary before it gets dumped to a file. Implementations of this hook can insert additional data into this dictionary.

Example:

def on_save_checkpoint(self, checkpoint):
    # 99% of use cases you don't need to implement this method
    checkpoint['something_cool_i_want_to_save'] = my_cool_pickable_object

Note

Lightning saves all aspects of training (epoch, global step, etc…) including amp scaling. There is no need for you to store anything about training.

on_test_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop after the batch.

Parameters

• outputs – The outputs of test_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_end() → None

Called at the end of testing.

on_test_epoch_end() → None

Called in the test loop at the very end of the epoch.

on_test_epoch_start() → None

Called in the test loop at the very beginning of the epoch.

on_test_model_eval() → None

Sets the model to eval during the test loop.

on_test_model_train() → None

Sets the model to train during the test loop.

on_test_start() → None

Called at the beginning of testing.

on_train_batch_end(outputs: Union[Tensor, Dict[str, Any]], batch: Any, batch_idx: int) → None

Called in the training loop after the batch.

Parameters

• outputs – The outputs of training_step_end(training_step(x))

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_batch_start(batch: Any, batch_idx: int) → Optional[int]

Called in the training loop before anything happens for that batch.

If you return -1 here, you will skip training for the rest of the current epoch.

Parameters

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_end() → None

Called at the end of training before logger experiment is closed.

on_train_epoch_end() → None

Called in the training loop at the very end of the epoch.

To access all batch outputs at the end of the epoch, either:

1. Implement training_epoch_end in the LightningModule OR

2. Cache data across steps on the attribute(s) of the LightningModule and access them in this hook

on_train_epoch_start() → None

Called in the training loop at the very beginning of the epoch.

on_train_start() → None

Called at the beginning of training after sanity check.

on_validation_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop after the batch.

Parameters

• outputs – The outputs of validation_step_end(validation_step(x))

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_end() → None

Called at the end of validation.

on_validation_epoch_end() → None

Called in the validation loop at the very end of the epoch.

on_validation_epoch_start() → None

Called in the validation loop at the very beginning of the epoch.

on_validation_model_eval() → None

Sets the model to eval during the val loop.

on_validation_model_train() → None

Sets the model to train during the val loop.

on_validation_start() → None

Called at the beginning of validation.

optimizer_step(epoch: int, batch_idx: int, optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int = 0, optimizer_closure: Optional[Callable[[], Any]] = None, on_tpu: bool = False, using_native_amp: bool = False, using_lbfgs: bool = False) → None

Override this method to adjust the default way the Trainer calls each optimizer.

By default, Lightning calls step() and zero_grad() as shown in the example once per optimizer. This method (and zero_grad()) won’t be called during the accumulation phase when Trainer(accumulate_grad_batches != 1). Overriding this hook has no benefit with manual optimization.

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers, this indexes into that list.

• optimizer_closure – The optimizer closure. This closure must be executed as it includes the calls to training_step(), optimizer.zero_grad(), and backward().

• on_tpu – True if TPU backward is required

• using_native_amp – True if using native amp

• using_lbfgs – True if the matching optimizer is torch.optim.LBFGS

Examples:

# DEFAULT
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    optimizer.step(closure=optimizer_closure)

# Alternating schedule for optimizer steps (i.e.: GANs)
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    # update generator opt every step
    if optimizer_idx == 0:
        optimizer.step(closure=optimizer_closure)

    # update discriminator opt every 2 steps
    if optimizer_idx == 1:
        if (batch_idx + 1) % 2 == 0 :
            optimizer.step(closure=optimizer_closure)
        else:
            # call the closure by itself to run `training_step` + `backward` without an optimizer step
            optimizer_closure()

    # ...
    # add as many optimizers as you want

Here’s another example showing how to use this for more advanced things such as learning rate warm-up:

# learning rate warm-up
def optimizer_step(
    self,
    epoch,
    batch_idx,
    optimizer,
    optimizer_idx,
    optimizer_closure,
    on_tpu,
    using_native_amp,
    using_lbfgs,
):
    # update params
    optimizer.step(closure=optimizer_closure)

    # manually warm up lr without a scheduler
    if self.trainer.global_step < 500:
        lr_scale = min(1.0, float(self.trainer.global_step + 1) / 500.0)
        for pg in optimizer.param_groups:
            pg["lr"] = lr_scale * self.learning_rate

optimizer_zero_grad(epoch: int, batch_idx: int, optimizer: Optimizer, optimizer_idx: int) → None

Override this method to change the default behaviour of optimizer.zero_grad().

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers this indexes into that list.

Examples:

# DEFAULT
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad()

# Set gradients to `None` instead of zero to improve performance.
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad(set_to_none=True)

See torch.optim.Optimizer.zero_grad() for the explanation of the above example.

optimizers(use_pl_optimizer: bool = True) → Union[Optimizer, LightningOptimizer, List[Optimizer], List[LightningOptimizer]]

Returns the optimizer(s) that are being used during training. Useful for manual optimization.

Parameters

use_pl_optimizer – If True, will wrap the optimizer(s) in a LightningOptimizer for automatic handling of precision and profiling.

Returns

A single optimizer, or a list of optimizers in case multiple ones are present.

pairwise_correlations(loader: DataLoader, train=False)

Calculates correlation for entire batch from dataloader

Parameters

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

Returns

by default returns the average pairwise correlation in each dimension (for 2 views just the correlation)

parameters(recurse: bool = True) → Iterator[Parameter]

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Parameters

recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

Parameter – module parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for param in model.parameters():
>>>     print(type(param), param.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

predict_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for prediction.

It’s recommended that all data downloads and preparation happen in prepare_data().

• predict()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying prediction samples.

Note

In the case where you return multiple prediction dataloaders, the predict_step() will have an argument dataloader_idx which matches the order here.

predict_step(batch: Any, batch_idx: int, dataloader_idx: int = 0) → Any

Step function called during predict(). By default, it calls forward(). Override to add any processing logic.

The predict_step() is used to scale inference on multi-devices.

To prevent an OOM error, it is possible to use BasePredictionWriter callback to write the predictions to disk or database after each batch or on epoch end.

The BasePredictionWriter should be used while using a spawn based accelerator. This happens for Trainer(strategy="ddp_spawn") or training on 8 TPU cores with Trainer(accelerator="tpu", devices=8) as predictions won’t be returned.

Example

class MyModel(LightningModule):

    def predict_step(self, batch, batch_idx, dataloader_idx=0):
        return self(batch)

dm = ...
model = MyModel()
trainer = Trainer(accelerator="gpu", devices=2)
predictions = trainer.predict(model, dm)

Parameters

• batch – Current batch.

• batch_idx – Index of current batch.

• dataloader_idx – Index of the current dataloader.

Returns

Predicted output

prepare_data() → None

Use this to download and prepare data. Downloading and saving data with multiple processes (distributed settings) will result in corrupted data. Lightning ensures this method is called only within a single process, so you can safely add your downloading logic within.

Warning

DO NOT set state to the model (use setup instead) since this is NOT called on every device

Example:

def prepare_data(self):
    # good
    download_data()
    tokenize()
    etc()

    # bad
    self.split = data_split
    self.some_state = some_other_state()

In a distributed environment, prepare_data can be called in two ways (using prepare_data_per_node)

1. Once per node. This is the default and is only called on LOCAL_RANK=0.

2. Once in total. Only called on GLOBAL_RANK=0.

Example:

# DEFAULT
# called once per node on LOCAL_RANK=0 of that node
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = True


# call on GLOBAL_RANK=0 (great for shared file systems)
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = False

This is called before requesting the dataloaders:

model.prepare_data()
initialize_distributed()
model.setup(stage)
model.train_dataloader()
model.val_dataloader()
model.test_dataloader()
model.predict_dataloader()

print(*args: Any, **kwargs: Any) → None

Prints only from process 0. Use this in any distributed mode to log only once.

Parameters

• *args – The thing to print. The same as for Python’s built-in print function.

• **kwargs – The same as for Python’s built-in print function.

Example:

def forward(self, x):
    self.print(x, 'in forward')

register_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

This function is deprecated in favor of register_full_backward_hook() and the behavior of this function will change in future versions.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_buffer(name: str, tensor: Optional[Tensor], persistent: bool = True) → None

Adds a buffer to the module.

This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the module’s state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module’s state_dict.

Buffers can be accessed as attributes using given names.

Parameters

• name (str) – name of the buffer. The buffer can be accessed from this module using the given name

• tensor (Tensor or None) – buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module’s state_dict.

• persistent (bool) – whether the buffer is part of this module’s state_dict.

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> self.register_buffer('running_mean', torch.zeros(num_features))

register_forward_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output. It should have the following signature:

hook(module, input, output) -> None or modified output

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward() is called.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_forward_pre_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked. It should have the following signature:

hook(module, input) -> None or modified input

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned(unless that value is already a tuple).

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_full_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

The hook will be called every time the gradients with respect to module inputs are computed. The hook should have the following signature:

hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module’s forward function.

Warning

Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_load_state_dict_post_hook(hook)

Registers a post hook to be run after module’s load_state_dict is called.

It should have the following signature::

hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict() with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearning out both missing and unexpected keys will avoid an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_module(name: str, module: Optional[Module]) → None

Alias for add_module().

register_parameter(name: str, param: Optional[Parameter]) → None

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Parameters

• name (str) – name of the parameter. The parameter can be accessed from this module using the given name

• param (Parameter or None) – parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module’s state_dict.

requires_grad_(requires_grad: bool = True) → T

Change if autograd should record operations on parameters in this module.

This method sets the parameters’ requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See Locally disabling gradient computation for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Parameters

requires_grad (bool) – whether autograd should record operations on parameters in this module. Default: True.

Returns

self

Return type

Module

save_hyperparameters(*args: Any, ignore: Optional[Union[Sequence[str], str]] = None, frame: Optional[frame] = None, logger: bool = True) → None

Save arguments to hparams attribute.

Parameters

• args – single object of dict, NameSpace or OmegaConf or string names or arguments from class __init__

• ignore – an argument name or a list of argument names from class __init__ to be ignored

• frame – a frame object. Default is None

• logger – Whether to send the hyperparameters to the logger. Default: True

Example::

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # manually assign arguments
...         self.save_hyperparameters('arg1', 'arg3')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

>>> class AutomaticArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # equivalent automatic
...         self.save_hyperparameters()
...     def forward(self, *args, **kwargs):
...         ...
>>> model = AutomaticArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg2": abc
"arg3": 3.14

>>> class SingleArgModel(HyperparametersMixin):
...     def __init__(self, params):
...         super().__init__()
...         # manually assign single argument
...         self.save_hyperparameters(params)
...     def forward(self, *args, **kwargs):
...         ...
>>> model = SingleArgModel(Namespace(p1=1, p2='abc', p3=3.14))
>>> model.hparams
"p1": 1
"p2": abc
"p3": 3.14

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # pass argument(s) to ignore as a string or in a list
...         self.save_hyperparameters(ignore='arg2')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

score(loader: DataLoader, **kwargs)

Returns average correlation in each dimension (averages over all pairs for multiview)

Parameters

• **kwargs –

• loader – a dataloader that matches the structure of that used for training

• train – whether to fit final linear transformation

set_extra_state(state: Any)

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Parameters

state (dict) – Extra state from the state_dict

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters

**params (dict) – Estimator parameters.

Returns

self – Estimator instance.

Return type

estimator instance

setup(stage: str) → None

Called at the beginning of fit (train + validate), validate, test, or predict. This is a good hook when you need to build models dynamically or adjust something about them. This hook is called on every process when using DDP.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

Example:

class LitModel(...):
    def __init__(self):
        self.l1 = None

    def prepare_data(self):
        download_data()
        tokenize()

        # don't do this
        self.something = else

    def setup(self, stage):
        data = load_data(...)
        self.l1 = nn.Linear(28, data.num_classes)

share_memory() → T

See torch.Tensor.share_memory_()

state_dict(*args, destination=None, prefix='', keep_vars=False)

Returns a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

Note

The returned object is a shallow copy. It contains references to the module’s parameters and buffers.

Warning

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

Warning

Please avoid the use of argument destination as it is not designed for end-users.

Parameters

• destination (dict, optional) – If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned. Default: None.

• prefix (str, optional) – a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.

• keep_vars (bool, optional) – by default the Tensor s returned in the state dict are detached from autograd. If it’s set to True, detaching will not be performed. Default: False.

Returns

a dictionary containing a whole state of the module

Return type

dict

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> module.state_dict().keys()
['bias', 'weight']

tbptt_split_batch(batch: Any, split_size: int) → List[Any]

When using truncated backpropagation through time, each batch must be split along the time dimension. Lightning handles this by default, but for custom behavior override this function.

Parameters

• batch – Current batch

• split_size – The size of the split

Returns

List of batch splits. Each split will be passed to training_step() to enable truncated back propagation through time. The default implementation splits root level Tensors and Sequences at dim=1 (i.e. time dim). It assumes that each time dim is the same length.

Examples:

def tbptt_split_batch(self, batch, split_size):
    splits = []
    for t in range(0, time_dims[0], split_size):
        batch_split = []
        for i, x in enumerate(batch):
            if isinstance(x, torch.Tensor):
                split_x = x[:, t:t + split_size]
            elif isinstance(x, collections.abc.Sequence):
                split_x = [None] * len(x)
                for batch_idx in range(len(x)):
                  split_x[batch_idx] = x[batch_idx][t:t + split_size]
            batch_split.append(split_x)
        splits.append(batch_split)
    return splits

Note

Called in the training loop after on_train_batch_start() if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0. Each returned batch split is passed separately to training_step().

teardown(stage: str) → None

Called at the end of fit (train + validate), validate, test, or predict.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

test_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for testing.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• test()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying testing samples.

Example:

def test_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def test_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a test dataset and a test_step(), you don’t need to implement this method.

Note

In the case where you return multiple test dataloaders, the test_step() will have an argument dataloader_idx which matches the order here.

test_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of a test epoch with the output of all test steps.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

outputs – List of outputs you defined in test_step_end(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader

Returns

None

Note

If you didn’t define a test_step(), this won’t be called.

Examples

With a single dataloader:

def test_epoch_end(self, outputs):
    # do something with the outputs of all test batches
    all_test_preds = test_step_outputs.predictions

    some_result = calc_all_results(all_test_preds)
    self.log(some_result)

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each test step for that dataloader.

def test_epoch_end(self, outputs):
    final_value = 0
    for dataloader_outputs in outputs:
        for test_step_out in dataloader_outputs:
            # do something
            final_value += test_step_out

    self.log("final_metric", final_value)

test_step(batch, batch_idx)

Operates on a single batch of data from the test set. In this step you’d normally generate examples or calculate anything of interest such as accuracy.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_id – The index of the dataloader that produced this batch. (only if multiple test dataloaders used).

Returns

Any of.

• Any object or value

• None - Testing will skip to the next batch

# if you have one test dataloader:
def test_step(self, batch, batch_idx):
    ...


# if you have multiple test dataloaders:
def test_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single test dataset
def test_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    test_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'test_loss': loss, 'test_acc': test_acc})

If you pass in multiple test dataloaders, test_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple test dataloaders
def test_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to test you don’t need to implement this method.

Note

When the test_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of the test epoch, the model goes back to training mode and gradients are enabled.

test_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when testing with DP because test_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [test_step(sub_batch) for sub_batch in sub_batches]
test_step_end(step_output)

Parameters

step_output – What you return in test_step() for each batch part.

Returns

None or anything

# WITHOUT test_step_end
# if used in DP, this batch is 1/num_gpus large
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    loss = self.softmax(out)
    self.log("test_loss", loss)


# --------------
# with test_step_end to do softmax over the full batch
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return out


def test_step_end(self, output_results):
    # this out is now the full size of the batch
    all_test_step_outs = output_results.out
    loss = nce_loss(all_test_step_outs)
    self.log("test_loss", loss)

See also

See the Multi GPU Training guide for more details.

to(*args: Any, **kwargs: Any) → Self

See torch.nn.Module.to().

to_empty(*, device: Union[str, device]) → T

Moves the parameters and buffers to the specified device without copying storage.

Parameters

device (torch.device) – The desired device of the parameters and buffers in this module.

Returns

self

Return type

Module

to_onnx(file_path: Union[str, Path], input_sample: Optional[Any] = None, **kwargs: Any) → None

Saves the model in ONNX format.

Parameters

• file_path – The path of the file the onnx model should be saved to.

• input_sample – An input for tracing. Default: None (Use self.example_input_array)

• **kwargs – Will be passed to torch.onnx.export function.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))

>>> with tempfile.NamedTemporaryFile(suffix='.onnx', delete=False) as tmpfile:
...     model = SimpleModel()
...     input_sample = torch.randn((1, 64))
...     model.to_onnx(tmpfile.name, input_sample, export_params=True)
...     os.path.isfile(tmpfile.name)
True

to_torchscript(file_path: Optional[Union[str, Path]] = None, method: Optional[str] = 'script', example_inputs: Optional[Any] = None, **kwargs: Any) → Union[ScriptModule, Dict[str, ScriptModule]]

By default compiles the whole model to a ScriptModule. If you want to use tracing, please provided the argument method='trace' and make sure that either the example_inputs argument is provided, or the model has example_input_array set. If you would like to customize the modules that are scripted you should override this method. In case you want to return multiple modules, we recommend using a dictionary.

Parameters

• file_path – Path where to save the torchscript. Default: None (no file saved).

• method – Whether to use TorchScript’s script or trace method. Default: ‘script’

• example_inputs – An input to be used to do tracing when method is set to ‘trace’. Default: None (uses example_input_array)

• **kwargs – Additional arguments that will be passed to the torch.jit.script() or torch.jit.trace() function.

Note

• Requires the implementation of the forward() method.

• The exported script will be set to evaluation mode.

• It is recommended that you install the latest supported version of PyTorch to use this feature without limitations. See also the torch.jit documentation for supported features.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))
...
>>> model = SimpleModel()
>>> model.to_torchscript(file_path="model.pt")  
>>> os.path.isfile("model.pt")  
>>> torch.jit.save(model.to_torchscript(file_path="model_trace.pt", method='trace', 
...                                     example_inputs=torch.randn(1, 64)))  
>>> os.path.isfile("model_trace.pt")  
True

Returns

This LightningModule as a torchscript, regardless of whether file_path is defined or not.

toggle_optimizer(optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int) → None

Makes sure only the gradients of the current optimizer’s parameters are calculated in the training step to prevent dangling gradients in multiple-optimizer setup.

This is only called automatically when automatic optimization is enabled and multiple optimizers are used. It works with untoggle_optimizer() to make sure param_requires_grad_state is properly reset.

Parameters

• optimizer – The optimizer to toggle.

• optimizer_idx – The index of the optimizer to toggle.

train(mode: bool = True) → T

Sets the module in training mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

Parameters

mode (bool) – whether to set training mode (True) or evaluation mode (False). Default: True.

Returns

self

Return type

Module

train_dataloader() → Union[DataLoader, Sequence[DataLoader], Sequence[Sequence[DataLoader]], Sequence[Dict[str, DataLoader]], Dict[str, DataLoader], Dict[str, Dict[str, DataLoader]], Dict[str, Sequence[DataLoader]]]

Implement one or more PyTorch DataLoaders for training.

Returns

A collection of torch.utils.data.DataLoader specifying training samples. In the case of multiple dataloaders, please see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• fit()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Example:

# single dataloader
def train_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=True, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=True
    )
    return loader

# multiple dataloaders, return as list
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a list of tensors: [batch_mnist, batch_cifar]
    return [mnist_loader, cifar_loader]

# multiple dataloader, return as dict
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a dict of tensors: {'mnist': batch_mnist, 'cifar': batch_cifar}
    return {'mnist': mnist_loader, 'cifar': cifar_loader}

training_epoch_end(outputs: List[Union[Tensor, Dict[str, Any]]]) → None

Called at the end of the training epoch with the outputs of all training steps. Use this in case you need to do something with all the outputs returned by training_step().

# the pseudocode for these calls
train_outs = []
for train_batch in train_data:
    out = training_step(train_batch)
    train_outs.append(out)
training_epoch_end(train_outs)

Parameters

outputs – List of outputs you defined in training_step(). If there are multiple optimizers or when using truncated_bptt_steps > 0, the lists have the dimensions (n_batches, tbptt_steps, n_optimizers). Dimensions of length 1 are squeezed.

Returns

None

Note

If this method is not overridden, this won’t be called.

def training_epoch_end(self, training_step_outputs):
    # do something with all training_step outputs
    for out in training_step_outputs:
        ...

training_step(batch, batch_idx)

Here you compute and return the training loss and some additional metrics for e.g. the progress bar or logger.

Parameters

• batch (Tensor | (Tensor, …) | [Tensor, …]) – The output of your DataLoader. A tensor, tuple or list.

• batch_idx (int) – Integer displaying index of this batch

• optimizer_idx (int) – When using multiple optimizers, this argument will also be present.

• hiddens (Any) – Passed in if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0.

Returns

Any of.

• Tensor - The loss tensor

• dict - A dictionary. Can include any keys, but must include the key 'loss'

• None - Training will skip to the next batch. This is only for automatic optimization.

  This is not supported for multi-GPU, TPU, IPU, or DeepSpeed.

In this step you’d normally do the forward pass and calculate the loss for a batch. You can also do fancier things like multiple forward passes or something model specific.

Example:

def training_step(self, batch, batch_idx):
    x, y, z = batch
    out = self.encoder(x)
    loss = self.loss(out, x)
    return loss

If you define multiple optimizers, this step will be called with an additional optimizer_idx parameter.

# Multiple optimizers (e.g.: GANs)
def training_step(self, batch, batch_idx, optimizer_idx):
    if optimizer_idx == 0:
        # do training_step with encoder
        ...
    if optimizer_idx == 1:
        # do training_step with decoder
        ...

If you add truncated back propagation through time you will also get an additional argument with the hidden states of the previous step.

# Truncated back-propagation through time
def training_step(self, batch, batch_idx, hiddens):
    # hiddens are the hidden states from the previous truncated backprop step
    out, hiddens = self.lstm(data, hiddens)
    loss = ...
    return {"loss": loss, "hiddens": hiddens}

Note

The loss value shown in the progress bar is smoothed (averaged) over the last values, so it differs from the actual loss returned in train/validation step.

Note

When accumulate_grad_batches > 1, the loss returned here will be automatically normalized by accumulate_grad_batches internally.

training_step_end(step_output: Union[Tensor, Dict[str, Any]]) → Union[Tensor, Dict[str, Any]]

Use this when training with dp because training_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [training_step(sub_batch) for sub_batch in sub_batches]
training_step_end(step_output)

Parameters

step_output – What you return in training_step for each batch part.

Returns

Anything

When using the DP strategy, only a portion of the batch is inside the training_step:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)

    # softmax uses only a portion of the batch in the denominator
    loss = self.softmax(out)
    loss = nce_loss(loss)
    return loss

If you wish to do something with all the parts of the batch, then use this method to do it:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return {"pred": out}


def training_step_end(self, training_step_outputs):
    gpu_0_pred = training_step_outputs[0]["pred"]
    gpu_1_pred = training_step_outputs[1]["pred"]
    gpu_n_pred = training_step_outputs[n]["pred"]

    # this softmax now uses the full batch
    loss = nce_loss([gpu_0_pred, gpu_1_pred, gpu_n_pred])
    return loss

See also

See the Multi GPU Training guide for more details.

transfer_batch_to_device(batch: Any, device: device, dataloader_idx: int) → Any

Override this hook if your DataLoader returns tensors wrapped in a custom data structure.

The data types listed below (and any arbitrary nesting of them) are supported out of the box:

• torch.Tensor or anything that implements .to(…)

• list

• dict

• tuple

For anything else, you need to define how the data is moved to the target device (CPU, GPU, TPU, …).

Note

This hook should only transfer the data and not modify it, nor should it move the data to any other device than the one passed in as argument (unless you know what you are doing). To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be transferred to a new device.

• device – The target device as defined in PyTorch.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A reference to the data on the new device.

Example:

def transfer_batch_to_device(self, batch, device, dataloader_idx):
    if isinstance(batch, CustomBatch):
        # move all tensors in your custom data structure to the device
        batch.samples = batch.samples.to(device)
        batch.targets = batch.targets.to(device)
    elif dataloader_idx == 0:
        # skip device transfer for the first dataloader or anything you wish
        pass
    else:
        batch = super().transfer_batch_to_device(data, device, dataloader_idx)
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• move_data_to_device()

• apply_to_collection()

transform(loader: DataLoader)

Parameters

loader – a dataloader that matches the structure of that used for training

Returns

transformed views

property truncated_bptt_steps: int

Enables Truncated Backpropagation Through Time in the Trainer when set to a positive integer.

It represents the number of times training_step() gets called before backpropagation. If this is > 0, the training_step() receives an additional argument hiddens and is expected to return a hidden state.

type(dst_type: Union[str, dtype]) → Self

See torch.nn.Module.type().

unfreeze() → None

Unfreeze all parameters for training.

model = MyLightningModule(...)
model.unfreeze()

untoggle_optimizer(optimizer_idx: int) → None

Resets the state of required gradients that were toggled with toggle_optimizer().

This is only called automatically when automatic optimization is enabled and multiple optimizers are used.

Parameters

optimizer_idx – The index of the optimizer to untoggle.

val_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for validation.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

It’s recommended that all data downloads and preparation happen in prepare_data().

• fit()

• validate()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying validation samples.

Examples:

def val_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False,
                    transform=transform, download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def val_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a validation dataset and a validation_step(), you don’t need to implement this method.

Note

In the case where you return multiple validation dataloaders, the validation_step() will have an argument dataloader_idx which matches the order here.

validation_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of the validation epoch with the outputs of all validation steps.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

outputs – List of outputs you defined in validation_step(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader.

Returns

None

Note

If you didn’t define a validation_step(), this won’t be called.

Examples

With a single dataloader:

def validation_epoch_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each validation step for that dataloader.

def validation_epoch_end(self, outputs):
    for dataloader_output_result in outputs:
        dataloader_outs = dataloader_output_result.dataloader_i_outputs

    self.log("final_metric", final_value)

validation_step(batch, batch_idx)

Operates on a single batch of data from the validation set. In this step you’d might generate examples or calculate anything of interest like accuracy.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_idx – The index of the dataloader that produced this batch. (only if multiple val dataloaders used)

Returns

• Any object or value

• None - Validation will skip to the next batch

# pseudocode of order
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    if defined("validation_step_end"):
        out = validation_step_end(out)
    val_outs.append(out)
val_outs = validation_epoch_end(val_outs)

# if you have one val dataloader:
def validation_step(self, batch, batch_idx):
    ...


# if you have multiple val dataloaders:
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single validation dataset
def validation_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    val_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'val_loss': loss, 'val_acc': val_acc})

If you pass in multiple val dataloaders, validation_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple validation dataloaders
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to validate you don’t need to implement this method.

Note

When the validation_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of validation, the model goes back to training mode and gradients are enabled.

validation_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when validating with dp because validation_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [validation_step(sub_batch) for sub_batch in sub_batches]
validation_step_end(step_output)

Parameters

step_output – What you return in validation_step() for each batch part.

Returns

None or anything

# WITHOUT validation_step_end
# if used in DP, this batch is 1/num_gpus large
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    loss = self.softmax(out)
    loss = nce_loss(loss)
    self.log("val_loss", loss)


# --------------
# with validation_step_end to do softmax over the full batch
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    return out


def validation_step_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

See also

See the Multi GPU Training guide for more details.

xpu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the XPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

zero_grad(set_to_none: bool = False) → None

Sets gradients of all model parameters to zero. See similar function under torch.optim.Optimizer for more context.

Parameters

set_to_none (bool) – instead of setting to zero, set the grads to None. See torch.optim.Optimizer.zero_grad() for details.

class cca_zoo.deepmodels.SplitAE(latent_dims: int, encoder=<class 'cca_zoo.deepmodels.architectures.Encoder'>, decoders=None, latent_dropout=0, recon_loss_type='mse', img_dim=None, **kwargs)

Bases: _BaseDeep, _GenerativeMixin

A class used to fit a Split Autoencoder model.

References

Ngiam, Jiquan, et al. “Multimodal deep learning.” ICML. 2011.

Parameters

• latent_dims – # latent dimensions

• encoder – list of encoder networks

• decoders – list of decoder networks

forward(views, **kwargs)

Forward method for the model. Outputs latent encoding for each view

Parameters

• views –

• kwargs –

Returns

loss(views, **kwargs)

Loss for model :param views: :param args: :param kwargs: :return:

configure_callbacks()

Configure model-specific callbacks. When the model gets attached, e.g., when .fit() or .test() gets called, the list or a callback returned here will be merged with the list of callbacks passed to the Trainer’s callbacks argument. If a callback returned here has the same type as one or several callbacks already present in the Trainer’s callbacks list, it will take priority and replace them. In addition, Lightning will make sure ModelCheckpoint callbacks run last.

Returns

A callback or a list of callbacks which will extend the list of callbacks in the Trainer.

Example:

def configure_callbacks(self):
    early_stop = EarlyStopping(monitor="val_acc", mode="max")
    checkpoint = ModelCheckpoint(monitor="val_loss")
    return [early_stop, checkpoint]

add_module(name: str, module: Optional[Module]) → None

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Parameters

• name (str) – name of the child module. The child module can be accessed from this module using the given name

• module (Module) – child module to be added to the module.

all_gather(data: Union[Tensor, Dict, List, Tuple], group: Optional[Any] = None, sync_grads: bool = False) → Union[Tensor, Dict, List, Tuple]

Allows users to call self.all_gather() from the LightningModule, thus making the all_gather operation accelerator agnostic. all_gather is a function provided by accelerators to gather a tensor from several distributed processes.

Parameters

• data – int, float, tensor of shape (batch, …), or a (possibly nested) collection thereof.

• group – the process group to gather results from. Defaults to all processes (world)

• sync_grads – flag that allows users to synchronize gradients for the all_gather operation

Returns

A tensor of shape (world_size, batch, …), or if the input was a collection the output will also be a collection with tensors of this shape.

apply(fn: Callable[[Module], None]) → T

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also torch.nn.init).

Parameters

fn (Module -> None) – function to be applied to each submodule

Returns

self

Return type

Module

Example:

>>> @torch.no_grad()
>>> def init_weights(m):
>>>     print(m)
>>>     if type(m) == nn.Linear:
>>>         m.weight.fill_(1.0)
>>>         print(m.weight)
>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

property automatic_optimization: bool

If set to False you are responsible for calling .backward(), .step(), .zero_grad().

backward(loss: Tensor, optimizer: Optional[Steppable], optimizer_idx: Optional[int], *args: Any, **kwargs: Any) → None

Called to perform backward on the loss returned in training_step(). Override this hook with your own implementation if you need to.

Parameters

• loss – The loss tensor returned by training_step(). If gradient accumulation is used, the loss here holds the normalized value (scaled by 1 / accumulation steps).

• optimizer – Current optimizer being used. None if using manual optimization.

• optimizer_idx – Index of the current optimizer being used. None if using manual optimization.

Example:

def backward(self, loss, optimizer, optimizer_idx):
    loss.backward()

bfloat16() → T

Casts all floating point parameters and buffers to bfloat16 datatype.

Note

This method modifies the module in-place.

Returns

self

Return type

Module

buffers(recurse: bool = True) → Iterator[Tensor]

Returns an iterator over module buffers.

Parameters

recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

torch.Tensor – module buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for buf in model.buffers():
>>>     print(type(buf), buf.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

children() → Iterator[Module]

Returns an iterator over immediate children modules.

Yields

Module – a child module

clip_gradients(optimizer: Optimizer, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Handles gradient clipping internally.

Note

Do not override this method. If you want to customize gradient clipping, consider using configure_gradient_clipping() method.

Parameters

• optimizer – Current optimizer being used.

• gradient_clip_val – The value at which to clip gradients.

• gradient_clip_algorithm – The gradient clipping algorithm to use. Pass gradient_clip_algorithm="value" to clip by value, and gradient_clip_algorithm="norm" to clip by norm.

configure_gradient_clipping(optimizer: Optimizer, optimizer_idx: int, gradient_clip_val: Optional[Union[int, float]] = None, gradient_clip_algorithm: Optional[str] = None) → None

Perform gradient clipping for the optimizer parameters. Called before optimizer_step().

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

• gradient_clip_val – The value at which to clip gradients. By default value passed in Trainer will be available here.

• gradient_clip_algorithm – The gradient clipping algorithm to use. By default value passed in Trainer will be available here.

Example:

# Perform gradient clipping on gradients associated with discriminator (optimizer_idx=1) in GAN
def configure_gradient_clipping(self, optimizer, optimizer_idx, gradient_clip_val, gradient_clip_algorithm):
    if optimizer_idx == 1:
        # Lightning will handle the gradient clipping
        self.clip_gradients(
            optimizer,
            gradient_clip_val=gradient_clip_val,
            gradient_clip_algorithm=gradient_clip_algorithm
        )
    else:
        # implement your own custom logic to clip gradients for generator (optimizer_idx=0)

configure_optimizers()

Collects learnable parameters and configures the optimizer and learning rate scheduler. :returns: two lists containing the optimizer and the scheduler. :rtype: Tuple[List, List]

configure_sharded_model() → None

Hook to create modules in a distributed aware context. This is useful for when using sharded plugins, where we’d like to shard the model instantly, which is useful for extremely large models which can save memory and initialization time.

This hook is called during each of fit/val/test/predict stages in the same process, so ensure that implementation of this hook is idempotent.

cpu() → Self

See torch.nn.Module.cpu().

cuda(device: Optional[Union[device, int]] = None) → Self

Moves all model parameters and buffers to the GPU. This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized.

Parameters

device – If specified, all parameters will be copied to that device. If None, the current CUDA device index will be used.

Returns

self

Return type

Module

property current_epoch: int

The current epoch in the Trainer, or 0 if not attached.

double() → Self

See torch.nn.Module.double().

eval() → T

Sets the module in evaluation mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False).

See Locally disabling gradient computation for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns

self

Return type

Module

property example_input_array: Optional[Union[Tensor, Tuple, Dict]]

The example input array is a specification of what the module can consume in the forward() method. The return type is interpreted as follows:

• Single tensor: It is assumed the model takes a single argument, i.e., model.forward(model.example_input_array)

• Tuple: The input array should be interpreted as a sequence of positional arguments, i.e., model.forward(*model.example_input_array)

• Dict: The input array represents named keyword arguments, i.e., model.forward(**model.example_input_array)

extra_repr() → str

Set the extra representation of the module

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float() → Self

See torch.nn.Module.float().

freeze() → None

Freeze all params for inference.

Example:

model = MyLightningModule(...)
model.freeze()

get_buffer(target: str) → Tensor

Returns the buffer given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The buffer referenced by target

Return type

torch.Tensor

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state() → Any

Returns any extra state to include in the module’s state_dict. Implement this and a corresponding set_extra_state() for your module if you need to store extra state. This function is called when building the module’s state_dict().

Note that extra state should be pickleable to ensure working serialization of the state_dict. We only provide provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns

Any extra state to store in the module’s state_dict

Return type

object

get_parameter(target: str) → Parameter

Returns the parameter given by target if it exists, otherwise throws an error.

See the docstring for get_submodule for a more detailed explanation of this method’s functionality as well as how to correctly specify target.

Parameters

target – The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns

The Parameter referenced by target

Return type

torch.nn.Parameter

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(target: str) → Module

Returns the submodule given by target if it exists, otherwise throws an error.

For example, let’s say you have an nn.Module A that looks like this:

A(
    (net_b): Module(
        (net_c): Module(
            (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2))
        )
        (linear): Linear(in_features=100, out_features=200, bias=True)
    )
)

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Parameters

target – The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns

The submodule referenced by target

Return type

torch.nn.Module

Raises

AttributeError – If the target string references an invalid path or resolves to something that is not an nn.Module

property global_rank: int

The index of the current process across all nodes and devices.

property global_step: int

Total training batches seen across all epochs.

If no Trainer is attached, this propery is 0.

half() → Self

See torch.nn.Module.half().

property hparams: Union[AttributeDict, MutableMapping]

The collection of hyperparameters saved with save_hyperparameters(). It is mutable by the user. For the frozen set of initial hyperparameters, use hparams_initial.

Returns

Mutable hyperparameters dictionary

property hparams_initial: AttributeDict

The collection of hyperparameters saved with save_hyperparameters(). These contents are read-only. Manual updates to the saved hyperparameters can instead be performed through hparams.

Returns

immutable initial hyperparameters

Return type

AttributeDict

ipu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the IPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on IPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

classmethod load_from_checkpoint(checkpoint_path: Union[str, Path, IO], map_location: Optional[Union[device, str, int, Callable[[Union[device, str, int]], Union[device, str, int]], Dict[Union[device, str, int], Union[device, str, int]]]] = None, hparams_file: Optional[Union[str, Path]] = None, strict: bool = True, **kwargs: Any) → Self

Primary way of loading a model from a checkpoint. When Lightning saves a checkpoint it stores the arguments passed to __init__ in the checkpoint under "hyper_parameters".

Any arguments specified through **kwargs will override args stored in "hyper_parameters".

Parameters

• checkpoint_path – Path to checkpoint. This can also be a URL, or file-like object

• map_location – If your checkpoint saved a GPU model and you now load on CPUs or a different number of GPUs, use this to map to the new setup. The behaviour is the same as in torch.load().

• hparams_file –

  Optional path to a .yaml or .csv file with hierarchical structure as in this example:

  drop_prob: 0.2
  dataloader:
      batch_size: 32
  

  You most likely won’t need this since Lightning will always save the hyperparameters to the checkpoint. However, if your checkpoint weights don’t have the hyperparameters saved, use this method to pass in a .yaml file with the hparams you’d like to use. These will be converted into a dict and passed into your LightningModule for use.

  If your model’s hparams argument is Namespace and .yaml file has hierarchical structure, you need to refactor your model to treat hparams as dict.

• strict – Whether to strictly enforce that the keys in checkpoint_path match the keys returned by this module’s state dict.

• **kwargs – Any extra keyword args needed to init the model. Can also be used to override saved hyperparameter values.

Returns

LightningModule instance with loaded weights and hyperparameters (if available).

Note

load_from_checkpoint is a class method. You should use your LightningModule class to call it instead of the LightningModule instance.

Example:

# load weights without mapping ...
model = MyLightningModule.load_from_checkpoint('path/to/checkpoint.ckpt')

# or load weights mapping all weights from GPU 1 to GPU 0 ...
map_location = {'cuda:1':'cuda:0'}
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    map_location=map_location
)

# or load weights and hyperparameters from separate files.
model = MyLightningModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    hparams_file='/path/to/hparams_file.yaml'
)

# override some of the params with new values
model = MyLightningModule.load_from_checkpoint(
    PATH,
    num_layers=128,
    pretrained_ckpt_path=NEW_PATH,
)

# predict
pretrained_model.eval()
pretrained_model.freeze()
y_hat = pretrained_model(x)

load_state_dict(state_dict: Mapping[str, Any], strict: bool = True)

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Parameters

• state_dict (dict) – a dict containing parameters and persistent buffers.

• strict (bool, optional) – whether to strictly enforce that the keys in state_dict match the keys returned by this module’s state_dict() function. Default: True

Returns

• missing_keys is a list of str containing the missing keys

• unexpected_keys is a list of str containing the unexpected keys

Return type

NamedTuple with missing_keys and unexpected_keys fields

Note

If a parameter or buffer is registered as None and its corresponding key exists in state_dict, load_state_dict() will raise a RuntimeError.

property local_rank: int

The index of the current process within a single node.

log(name: str, value: Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, metric_attribute: Optional[str] = None, rank_zero_only: bool = False) → None

Log a key, value pair.

Example:

self.log('train_loss', loss)

The default behavior per hook is documented here: extensions/logging:Automatic Logging.

Parameters

• name – key to log.

• value – value to log. Can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress bar.

• logger – if True logs to the logger.

• on_step – if True logs at this step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto detach the graph.

• sync_dist – if True, reduces the metric across devices. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the DDP group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple dataloaders). If False, user needs to give unique names for each dataloader to not mix the values.

• batch_size – Current batch_size. This will be directly inferred from the loaded batch, but for some data structures you might need to explicitly provide it.

• metric_attribute – To restore the metric state, Lightning requires the reference of the torchmetrics.Metric in your model. This is found automatically if it is a model attribute.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_dict(dictionary: Mapping[str, Union[Metric, Tensor, int, float, Mapping[str, Union[Metric, Tensor, int, float]]]], prog_bar: bool = False, logger: bool = True, on_step: Optional[bool] = None, on_epoch: Optional[bool] = None, reduce_fx: Union[str, Callable] = 'mean', enable_graph: bool = False, sync_dist: bool = False, sync_dist_group: Optional[Any] = None, add_dataloader_idx: bool = True, batch_size: Optional[int] = None, rank_zero_only: bool = False) → None

Log a dictionary of values at once.

Example:

values = {'loss': loss, 'acc': acc, ..., 'metric_n': metric_n}
self.log_dict(values)

Parameters

• dictionary – key value pairs. The values can be a float, Tensor, Metric, or a dictionary of the former.

• prog_bar – if True logs to the progress base.

• logger – if True logs to the logger.

• on_step – if True logs at this step. None auto-logs for training_step but not validation/test_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• on_epoch – if True logs epoch accumulated metrics. None auto-logs for val/test step but not training_step. The default value is determined by the hook. See extensions/logging:Automatic Logging for details.

• reduce_fx – reduction function over step values for end of epoch. torch.mean() by default.

• enable_graph – if True, will not auto-detach the graph

• sync_dist – if True, reduces the metric across GPUs/TPUs. Use with care as this may lead to a significant communication overhead.

• sync_dist_group – the ddp group to sync across.

• add_dataloader_idx – if True, appends the index of the current dataloader to the name (when using multiple). If False, user needs to give unique names for each dataloader to not mix values.

• batch_size – Current batch size. This will be directly inferred from the loaded batch, but some data structures might need to explicitly provide it.

• rank_zero_only – Whether the value will be logged only on rank 0. This will prevent synchronization which would produce a deadlock as not all processes would perform this log call.

log_grad_norm(grad_norm_dict: Dict[str, float]) → None

Override this method to change the default behaviour of log_grad_norm.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

grad_norm_dict – Dictionary containing current grad norm metrics

Example:

# DEFAULT
def log_grad_norm(self, grad_norm_dict):
    self.log_dict(grad_norm_dict, on_step=True, on_epoch=True, prog_bar=False, logger=True)

property logger: Optional[Logger]

Reference to the logger object in the Trainer.

property loggers: List[Logger]

Reference to the list of loggers in the Trainer.

lr_scheduler_step(scheduler: Union[_LRScheduler, ReduceLROnPlateau], optimizer_idx: int, metric: Optional[Any]) → None

Override this method to adjust the default way the Trainer calls each scheduler. By default, Lightning calls step() and as shown in the example for each scheduler based on its interval.

Parameters

• scheduler – Learning rate scheduler.

• optimizer_idx – Index of the optimizer associated with this scheduler.

• metric – Value of the monitor used for schedulers like ReduceLROnPlateau.

Examples:

# DEFAULT
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    if metric is None:
        scheduler.step()
    else:
        scheduler.step(metric)

# Alternative way to update schedulers if it requires an epoch value
def lr_scheduler_step(self, scheduler, optimizer_idx, metric):
    scheduler.step(epoch=self.current_epoch)

lr_schedulers() → Union[None, List[Union[_LRScheduler, ReduceLROnPlateau]], _LRScheduler, ReduceLROnPlateau]

Returns the learning rate scheduler(s) that are being used during training. Useful for manual optimization.

Returns

A single scheduler, or a list of schedulers in case multiple ones are present, or None if no schedulers were returned in configure_optimizers().

manual_backward(loss: Tensor, *args: Any, **kwargs: Any) → None

Call this directly from your training_step() when doing optimizations manually. By using this, Lightning can ensure that all the proper scaling gets applied when using mixed precision.

See manual optimization for more examples.

Example:

def training_step(...):
    opt = self.optimizers()
    loss = ...
    opt.zero_grad()
    # automatically applies scaling, etc...
    self.manual_backward(loss)
    opt.step()

Parameters

• loss – The tensor on which to compute gradients. Must have a graph attached.

• *args – Additional positional arguments to be forwarded to backward()

• **kwargs – Additional keyword arguments to be forwarded to backward()

modules() → Iterator[Module]

Returns an iterator over all modules in the network.

Yields

Module – a module in the network

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.modules()):
...     print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Tensor]]

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Parameters

• prefix (str) – prefix to prepend to all buffer names.

• recurse (bool) – if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields

(str, torch.Tensor) – Tuple containing the name and buffer

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, buf in self.named_buffers():
>>>    if name in ['running_var']:
>>>        print(buf.size())

named_children() → Iterator[Tuple[str, Module]]

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields

(str, Module) – Tuple containing a name and child module

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, module in model.named_children():
>>>     if name in ['conv4', 'conv5']:
>>>         print(module)

named_modules(memo: Optional[Set[Module]] = None, prefix: str = '', remove_duplicate: bool = True)

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Parameters

• memo – a memo to store the set of modules already added to the result

• prefix – a prefix that will be added to the name of the module

• remove_duplicate – whether to remove the duplicated module instances in the result or not

Yields

(str, Module) – Tuple of name and module

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.named_modules()):
...     print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(prefix: str = '', recurse: bool = True) → Iterator[Tuple[str, Parameter]]

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Parameters

• prefix (str) – prefix to prepend to all parameter names.

• recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

(str, Parameter) – Tuple containing the name and parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for name, param in self.named_parameters():
>>>    if name in ['bias']:
>>>        print(param.size())

on_after_backward() → None

Called after loss.backward() and before optimizers are stepped.

Note

If using native AMP, the gradients will not be unscaled at this point. Use the on_before_optimizer_step if you need the unscaled gradients.

on_after_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch after it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_after_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = gpu_transforms(batch['x'])
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• on_before_batch_transfer()

• transfer_batch_to_device()

on_before_backward(loss: Tensor) → None

Called before loss.backward().

Parameters

loss – Loss divided by number of batches for gradient accumulation and scaled if using native AMP.

on_before_batch_transfer(batch: Any, dataloader_idx: int) → Any

Override to alter or apply batch augmentations to your batch before it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be altered or augmented.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A batch of data

Example:

def on_before_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = transforms(batch['x'])
    return batch

See also

• on_after_batch_transfer()

• transfer_batch_to_device()

on_before_optimizer_step(optimizer: Optimizer, optimizer_idx: int) → None

Called before optimizer.step().

If using gradient accumulation, the hook is called once the gradients have been accumulated. See: :paramref:`~pytorch_lightning.trainer.Trainer.accumulate_grad_batches`.

If using native AMP, the loss will be unscaled before calling this hook. See these docs for more information on the scaling of gradients.

If clipping gradients, the gradients will not have been clipped yet.

Parameters

• optimizer – Current optimizer being used.

• optimizer_idx – Index of the current optimizer being used.

Example:

def on_before_optimizer_step(self, optimizer, optimizer_idx):
    # example to inspect gradient information in tensorboard
    if self.trainer.global_step % 25 == 0:  # don't make the tf file huge
        for k, v in self.named_parameters():
            self.logger.experiment.add_histogram(
                tag=k, values=v.grad, global_step=self.trainer.global_step
            )

on_before_zero_grad(optimizer: Optimizer) → None

Called after training_step() and before optimizer.zero_grad().

Called in the training loop after taking an optimizer step and before zeroing grads. Good place to inspect weight information with weights updated.

This is where it is called:

for optimizer in optimizers:
    out = training_step(...)

    model.on_before_zero_grad(optimizer) # < ---- called here
    optimizer.zero_grad()

    backward()

Parameters

optimizer – The optimizer for which grads should be zeroed.

on_fit_end() → None

Called at the very end of fit.

If on DDP it is called on every process

on_fit_start() → None

Called at the very beginning of fit.

If on DDP it is called on every process

property on_gpu: bool

Returns True if this model is currently located on a GPU.

Useful to set flags around the LightningModule for different CPU vs GPU behavior.

on_load_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning to restore your model. If you saved something with on_save_checkpoint() this is your chance to restore this.

Parameters

checkpoint – Loaded checkpoint

Example:

def on_load_checkpoint(self, checkpoint):
    # 99% of the time you don't need to implement this method
    self.something_cool_i_want_to_save = checkpoint['something_cool_i_want_to_save']

Note

Lightning auto-restores global step, epoch, and train state including amp scaling. There is no need for you to restore anything regarding training.

on_predict_batch_end(outputs: Optional[Any], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop after the batch.

Parameters

• outputs – The outputs of predict_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the predict loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_predict_end() → None

Called at the end of predicting.

on_predict_epoch_end(results: List[Any]) → None

Called at the end of predicting.

on_predict_epoch_start() → None

Called at the beginning of predicting.

on_predict_model_eval() → None

Sets the model to eval during the predict loop.

on_predict_start() → None

Called at the beginning of predicting.

on_save_checkpoint(checkpoint: Dict[str, Any]) → None

Called by Lightning when saving a checkpoint to give you a chance to store anything else you might want to save.

Parameters

checkpoint – The full checkpoint dictionary before it gets dumped to a file. Implementations of this hook can insert additional data into this dictionary.

Example:

def on_save_checkpoint(self, checkpoint):
    # 99% of use cases you don't need to implement this method
    checkpoint['something_cool_i_want_to_save'] = my_cool_pickable_object

Note

Lightning saves all aspects of training (epoch, global step, etc…) including amp scaling. There is no need for you to store anything about training.

on_test_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop after the batch.

Parameters

• outputs – The outputs of test_step_end(test_step(x))

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the test loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the test DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_test_end() → None

Called at the end of testing.

on_test_epoch_end() → None

Called in the test loop at the very end of the epoch.

on_test_epoch_start() → None

Called in the test loop at the very beginning of the epoch.

on_test_model_eval() → None

Sets the model to eval during the test loop.

on_test_model_train() → None

Sets the model to train during the test loop.

on_test_start() → None

Called at the beginning of testing.

on_train_batch_end(outputs: Union[Tensor, Dict[str, Any]], batch: Any, batch_idx: int) → None

Called in the training loop after the batch.

Parameters

• outputs – The outputs of training_step_end(training_step(x))

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_batch_start(batch: Any, batch_idx: int) → Optional[int]

Called in the training loop before anything happens for that batch.

If you return -1 here, you will skip training for the rest of the current epoch.

Parameters

• batch – The batched data as it is returned by the training DataLoader.

• batch_idx – the index of the batch

on_train_end() → None

Called at the end of training before logger experiment is closed.

on_train_epoch_end() → None

Called in the training loop at the very end of the epoch.

To access all batch outputs at the end of the epoch, either:

1. Implement training_epoch_end in the LightningModule OR

2. Cache data across steps on the attribute(s) of the LightningModule and access them in this hook

on_train_epoch_start() → None

Called in the training loop at the very beginning of the epoch.

on_train_start() → None

Called at the beginning of training after sanity check.

on_validation_batch_end(outputs: Optional[Union[Tensor, Dict[str, Any]]], batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop after the batch.

Parameters

• outputs – The outputs of validation_step_end(validation_step(x))

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_batch_start(batch: Any, batch_idx: int, dataloader_idx: int) → None

Called in the validation loop before anything happens for that batch.

Parameters

• batch – The batched data as it is returned by the validation DataLoader.

• batch_idx – the index of the batch

• dataloader_idx – the index of the dataloader

on_validation_end() → None

Called at the end of validation.

on_validation_epoch_end() → None

Called in the validation loop at the very end of the epoch.

on_validation_epoch_start() → None

Called in the validation loop at the very beginning of the epoch.

on_validation_model_eval() → None

Sets the model to eval during the val loop.

on_validation_model_train() → None

Sets the model to train during the val loop.

on_validation_start() → None

Called at the beginning of validation.

optimizer_step(epoch: int, batch_idx: int, optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int = 0, optimizer_closure: Optional[Callable[[], Any]] = None, on_tpu: bool = False, using_native_amp: bool = False, using_lbfgs: bool = False) → None

Override this method to adjust the default way the Trainer calls each optimizer.

By default, Lightning calls step() and zero_grad() as shown in the example once per optimizer. This method (and zero_grad()) won’t be called during the accumulation phase when Trainer(accumulate_grad_batches != 1). Overriding this hook has no benefit with manual optimization.

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers, this indexes into that list.

• optimizer_closure – The optimizer closure. This closure must be executed as it includes the calls to training_step(), optimizer.zero_grad(), and backward().

• on_tpu – True if TPU backward is required

• using_native_amp – True if using native amp

• using_lbfgs – True if the matching optimizer is torch.optim.LBFGS

Examples:

# DEFAULT
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    optimizer.step(closure=optimizer_closure)

# Alternating schedule for optimizer steps (i.e.: GANs)
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx,
                   optimizer_closure, on_tpu, using_native_amp, using_lbfgs):
    # update generator opt every step
    if optimizer_idx == 0:
        optimizer.step(closure=optimizer_closure)

    # update discriminator opt every 2 steps
    if optimizer_idx == 1:
        if (batch_idx + 1) % 2 == 0 :
            optimizer.step(closure=optimizer_closure)
        else:
            # call the closure by itself to run `training_step` + `backward` without an optimizer step
            optimizer_closure()

    # ...
    # add as many optimizers as you want

Here’s another example showing how to use this for more advanced things such as learning rate warm-up:

# learning rate warm-up
def optimizer_step(
    self,
    epoch,
    batch_idx,
    optimizer,
    optimizer_idx,
    optimizer_closure,
    on_tpu,
    using_native_amp,
    using_lbfgs,
):
    # update params
    optimizer.step(closure=optimizer_closure)

    # manually warm up lr without a scheduler
    if self.trainer.global_step < 500:
        lr_scale = min(1.0, float(self.trainer.global_step + 1) / 500.0)
        for pg in optimizer.param_groups:
            pg["lr"] = lr_scale * self.learning_rate

optimizer_zero_grad(epoch: int, batch_idx: int, optimizer: Optimizer, optimizer_idx: int) → None

Override this method to change the default behaviour of optimizer.zero_grad().

Parameters

• epoch – Current epoch

• batch_idx – Index of current batch

• optimizer – A PyTorch optimizer

• optimizer_idx – If you used multiple optimizers this indexes into that list.

Examples:

# DEFAULT
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad()

# Set gradients to `None` instead of zero to improve performance.
def optimizer_zero_grad(self, epoch, batch_idx, optimizer, optimizer_idx):
    optimizer.zero_grad(set_to_none=True)

See torch.optim.Optimizer.zero_grad() for the explanation of the above example.

optimizers(use_pl_optimizer: bool = True) → Union[Optimizer, LightningOptimizer, List[Optimizer], List[LightningOptimizer]]

Returns the optimizer(s) that are being used during training. Useful for manual optimization.

Parameters

use_pl_optimizer – If True, will wrap the optimizer(s) in a LightningOptimizer for automatic handling of precision and profiling.

Returns

A single optimizer, or a list of optimizers in case multiple ones are present.

parameters(recurse: bool = True) → Iterator[Parameter]

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Parameters

recurse (bool) – if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields

Parameter – module parameter

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> for param in model.parameters():
>>>     print(type(param), param.size())
<class 'torch.Tensor'> (20L,)
<class 'torch.Tensor'> (20L, 1L, 5L, 5L)

predict_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for prediction.

It’s recommended that all data downloads and preparation happen in prepare_data().

• predict()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying prediction samples.

Note

In the case where you return multiple prediction dataloaders, the predict_step() will have an argument dataloader_idx which matches the order here.

predict_step(batch: Any, batch_idx: int, dataloader_idx: int = 0) → Any

Step function called during predict(). By default, it calls forward(). Override to add any processing logic.

The predict_step() is used to scale inference on multi-devices.

To prevent an OOM error, it is possible to use BasePredictionWriter callback to write the predictions to disk or database after each batch or on epoch end.

The BasePredictionWriter should be used while using a spawn based accelerator. This happens for Trainer(strategy="ddp_spawn") or training on 8 TPU cores with Trainer(accelerator="tpu", devices=8) as predictions won’t be returned.

Example

class MyModel(LightningModule):

    def predict_step(self, batch, batch_idx, dataloader_idx=0):
        return self(batch)

dm = ...
model = MyModel()
trainer = Trainer(accelerator="gpu", devices=2)
predictions = trainer.predict(model, dm)

Parameters

• batch – Current batch.

• batch_idx – Index of current batch.

• dataloader_idx – Index of the current dataloader.

Returns

Predicted output

prepare_data() → None

Use this to download and prepare data. Downloading and saving data with multiple processes (distributed settings) will result in corrupted data. Lightning ensures this method is called only within a single process, so you can safely add your downloading logic within.

Warning

DO NOT set state to the model (use setup instead) since this is NOT called on every device

Example:

def prepare_data(self):
    # good
    download_data()
    tokenize()
    etc()

    # bad
    self.split = data_split
    self.some_state = some_other_state()

In a distributed environment, prepare_data can be called in two ways (using prepare_data_per_node)

1. Once per node. This is the default and is only called on LOCAL_RANK=0.

2. Once in total. Only called on GLOBAL_RANK=0.

Example:

# DEFAULT
# called once per node on LOCAL_RANK=0 of that node
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = True


# call on GLOBAL_RANK=0 (great for shared file systems)
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = False

This is called before requesting the dataloaders:

model.prepare_data()
initialize_distributed()
model.setup(stage)
model.train_dataloader()
model.val_dataloader()
model.test_dataloader()
model.predict_dataloader()

print(*args: Any, **kwargs: Any) → None

Prints only from process 0. Use this in any distributed mode to log only once.

Parameters

• *args – The thing to print. The same as for Python’s built-in print function.

• **kwargs – The same as for Python’s built-in print function.

Example:

def forward(self, x):
    self.print(x, 'in forward')

register_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

This function is deprecated in favor of register_full_backward_hook() and the behavior of this function will change in future versions.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_buffer(name: str, tensor: Optional[Tensor], persistent: bool = True) → None

Adds a buffer to the module.

This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the module’s state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module’s state_dict.

Buffers can be accessed as attributes using given names.

Parameters

• name (str) – name of the buffer. The buffer can be accessed from this module using the given name

• tensor (Tensor or None) – buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module’s state_dict.

• persistent (bool) – whether the buffer is part of this module’s state_dict.

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> self.register_buffer('running_mean', torch.zeros(num_features))

register_forward_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output. It should have the following signature:

hook(module, input, output) -> None or modified output

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward() is called.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_forward_pre_hook(hook: Callable[[...], None]) → RemovableHandle

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked. It should have the following signature:

hook(module, input) -> None or modified input

The input contains only the positional arguments given to the module. Keyword arguments won’t be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned(unless that value is already a tuple).

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_full_backward_hook(hook: Callable[[Module, Union[Tuple[Tensor, ...], Tensor], Union[Tuple[Tensor, ...], Tensor]], Union[None, Tensor]]) → RemovableHandle

Registers a backward hook on the module.

The hook will be called every time the gradients with respect to module inputs are computed. The hook should have the following signature:

hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module’s forward function.

Warning

Modifying inputs or outputs inplace is not allowed when using backward hooks and will raise an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_load_state_dict_post_hook(hook)

Registers a post hook to be run after module’s load_state_dict is called.

It should have the following signature::

hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict() with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearning out both missing and unexpected keys will avoid an error.

Returns

a handle that can be used to remove the added hook by calling handle.remove()

Return type

torch.utils.hooks.RemovableHandle

register_module(name: str, module: Optional[Module]) → None

Alias for add_module().

register_parameter(name: str, param: Optional[Parameter]) → None

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Parameters

• name (str) – name of the parameter. The parameter can be accessed from this module using the given name

• param (Parameter or None) – parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module’s state_dict.

requires_grad_(requires_grad: bool = True) → T

Change if autograd should record operations on parameters in this module.

This method sets the parameters’ requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See Locally disabling gradient computation for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Parameters

requires_grad (bool) – whether autograd should record operations on parameters in this module. Default: True.

Returns

self

Return type

Module

save_hyperparameters(*args: Any, ignore: Optional[Union[Sequence[str], str]] = None, frame: Optional[frame] = None, logger: bool = True) → None

Save arguments to hparams attribute.

Parameters

• args – single object of dict, NameSpace or OmegaConf or string names or arguments from class __init__

• ignore – an argument name or a list of argument names from class __init__ to be ignored

• frame – a frame object. Default is None

• logger – Whether to send the hyperparameters to the logger. Default: True

Example::

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # manually assign arguments
...         self.save_hyperparameters('arg1', 'arg3')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

>>> class AutomaticArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # equivalent automatic
...         self.save_hyperparameters()
...     def forward(self, *args, **kwargs):
...         ...
>>> model = AutomaticArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg2": abc
"arg3": 3.14

>>> class SingleArgModel(HyperparametersMixin):
...     def __init__(self, params):
...         super().__init__()
...         # manually assign single argument
...         self.save_hyperparameters(params)
...     def forward(self, *args, **kwargs):
...         ...
>>> model = SingleArgModel(Namespace(p1=1, p2='abc', p3=3.14))
>>> model.hparams
"p1": 1
"p2": abc
"p3": 3.14

>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # pass argument(s) to ignore as a string or in a list
...         self.save_hyperparameters(ignore='arg2')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14

set_extra_state(state: Any)

This function is called from load_state_dict() to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state() for your module if you need to store extra state within its state_dict.

Parameters

state (dict) – Extra state from the state_dict

setup(stage: str) → None

Called at the beginning of fit (train + validate), validate, test, or predict. This is a good hook when you need to build models dynamically or adjust something about them. This hook is called on every process when using DDP.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

Example:

class LitModel(...):
    def __init__(self):
        self.l1 = None

    def prepare_data(self):
        download_data()
        tokenize()

        # don't do this
        self.something = else

    def setup(self, stage):
        data = load_data(...)
        self.l1 = nn.Linear(28, data.num_classes)

share_memory() → T

See torch.Tensor.share_memory_()

state_dict(*args, destination=None, prefix='', keep_vars=False)

Returns a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

Note

The returned object is a shallow copy. It contains references to the module’s parameters and buffers.

Warning

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

Warning

Please avoid the use of argument destination as it is not designed for end-users.

Parameters

• destination (dict, optional) – If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned. Default: None.

• prefix (str, optional) – a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.

• keep_vars (bool, optional) – by default the Tensor s returned in the state dict are detached from autograd. If it’s set to True, detaching will not be performed. Default: False.

Returns

a dictionary containing a whole state of the module

Return type

dict

Example:

>>> # xdoctest: +SKIP("undefined vars")
>>> module.state_dict().keys()
['bias', 'weight']

tbptt_split_batch(batch: Any, split_size: int) → List[Any]

When using truncated backpropagation through time, each batch must be split along the time dimension. Lightning handles this by default, but for custom behavior override this function.

Parameters

• batch – Current batch

• split_size – The size of the split

Returns

List of batch splits. Each split will be passed to training_step() to enable truncated back propagation through time. The default implementation splits root level Tensors and Sequences at dim=1 (i.e. time dim). It assumes that each time dim is the same length.

Examples:

def tbptt_split_batch(self, batch, split_size):
    splits = []
    for t in range(0, time_dims[0], split_size):
        batch_split = []
        for i, x in enumerate(batch):
            if isinstance(x, torch.Tensor):
                split_x = x[:, t:t + split_size]
            elif isinstance(x, collections.abc.Sequence):
                split_x = [None] * len(x)
                for batch_idx in range(len(x)):
                  split_x[batch_idx] = x[batch_idx][t:t + split_size]
            batch_split.append(split_x)
        splits.append(batch_split)
    return splits

Note

Called in the training loop after on_train_batch_start() if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0. Each returned batch split is passed separately to training_step().

teardown(stage: str) → None

Called at the end of fit (train + validate), validate, test, or predict.

Parameters

stage – either 'fit', 'validate', 'test', or 'predict'

test_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for testing.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• test()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying testing samples.

Example:

def test_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def test_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a test dataset and a test_step(), you don’t need to implement this method.

Note

In the case where you return multiple test dataloaders, the test_step() will have an argument dataloader_idx which matches the order here.

test_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of a test epoch with the output of all test steps.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

outputs – List of outputs you defined in test_step_end(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader

Returns

None

Note

If you didn’t define a test_step(), this won’t be called.

Examples

With a single dataloader:

def test_epoch_end(self, outputs):
    # do something with the outputs of all test batches
    all_test_preds = test_step_outputs.predictions

    some_result = calc_all_results(all_test_preds)
    self.log(some_result)

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each test step for that dataloader.

def test_epoch_end(self, outputs):
    final_value = 0
    for dataloader_outputs in outputs:
        for test_step_out in dataloader_outputs:
            # do something
            final_value += test_step_out

    self.log("final_metric", final_value)

test_step(batch, batch_idx)

Operates on a single batch of data from the test set. In this step you’d normally generate examples or calculate anything of interest such as accuracy.

# the pseudocode for these calls
test_outs = []
for test_batch in test_data:
    out = test_step(test_batch)
    test_outs.append(out)
test_epoch_end(test_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_id – The index of the dataloader that produced this batch. (only if multiple test dataloaders used).

Returns

Any of.

• Any object or value

• None - Testing will skip to the next batch

# if you have one test dataloader:
def test_step(self, batch, batch_idx):
    ...


# if you have multiple test dataloaders:
def test_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single test dataset
def test_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    test_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'test_loss': loss, 'test_acc': test_acc})

If you pass in multiple test dataloaders, test_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple test dataloaders
def test_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to test you don’t need to implement this method.

Note

When the test_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of the test epoch, the model goes back to training mode and gradients are enabled.

test_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when testing with DP because test_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [test_step(sub_batch) for sub_batch in sub_batches]
test_step_end(step_output)

Parameters

step_output – What you return in test_step() for each batch part.

Returns

None or anything

# WITHOUT test_step_end
# if used in DP, this batch is 1/num_gpus large
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    loss = self.softmax(out)
    self.log("test_loss", loss)


# --------------
# with test_step_end to do softmax over the full batch
def test_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return out


def test_step_end(self, output_results):
    # this out is now the full size of the batch
    all_test_step_outs = output_results.out
    loss = nce_loss(all_test_step_outs)
    self.log("test_loss", loss)

See also

See the Multi GPU Training guide for more details.

to(*args: Any, **kwargs: Any) → Self

See torch.nn.Module.to().

to_empty(*, device: Union[str, device]) → T

Moves the parameters and buffers to the specified device without copying storage.

Parameters

device (torch.device) – The desired device of the parameters and buffers in this module.

Returns

self

Return type

Module

to_onnx(file_path: Union[str, Path], input_sample: Optional[Any] = None, **kwargs: Any) → None

Saves the model in ONNX format.

Parameters

• file_path – The path of the file the onnx model should be saved to.

• input_sample – An input for tracing. Default: None (Use self.example_input_array)

• **kwargs – Will be passed to torch.onnx.export function.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))

>>> with tempfile.NamedTemporaryFile(suffix='.onnx', delete=False) as tmpfile:
...     model = SimpleModel()
...     input_sample = torch.randn((1, 64))
...     model.to_onnx(tmpfile.name, input_sample, export_params=True)
...     os.path.isfile(tmpfile.name)
True

to_torchscript(file_path: Optional[Union[str, Path]] = None, method: Optional[str] = 'script', example_inputs: Optional[Any] = None, **kwargs: Any) → Union[ScriptModule, Dict[str, ScriptModule]]

By default compiles the whole model to a ScriptModule. If you want to use tracing, please provided the argument method='trace' and make sure that either the example_inputs argument is provided, or the model has example_input_array set. If you would like to customize the modules that are scripted you should override this method. In case you want to return multiple modules, we recommend using a dictionary.

Parameters

• file_path – Path where to save the torchscript. Default: None (no file saved).

• method – Whether to use TorchScript’s script or trace method. Default: ‘script’

• example_inputs – An input to be used to do tracing when method is set to ‘trace’. Default: None (uses example_input_array)

• **kwargs – Additional arguments that will be passed to the torch.jit.script() or torch.jit.trace() function.

Note

• Requires the implementation of the forward() method.

• The exported script will be set to evaluation mode.

• It is recommended that you install the latest supported version of PyTorch to use this feature without limitations. See also the torch.jit documentation for supported features.

Example

>>> class SimpleModel(LightningModule):
...     def __init__(self):
...         super().__init__()
...         self.l1 = torch.nn.Linear(in_features=64, out_features=4)
...
...     def forward(self, x):
...         return torch.relu(self.l1(x.view(x.size(0), -1)))
...
>>> model = SimpleModel()
>>> model.to_torchscript(file_path="model.pt")  
>>> os.path.isfile("model.pt")  
>>> torch.jit.save(model.to_torchscript(file_path="model_trace.pt", method='trace', 
...                                     example_inputs=torch.randn(1, 64)))  
>>> os.path.isfile("model_trace.pt")  
True

Returns

This LightningModule as a torchscript, regardless of whether file_path is defined or not.

toggle_optimizer(optimizer: Union[Optimizer, LightningOptimizer], optimizer_idx: int) → None

Makes sure only the gradients of the current optimizer’s parameters are calculated in the training step to prevent dangling gradients in multiple-optimizer setup.

This is only called automatically when automatic optimization is enabled and multiple optimizers are used. It works with untoggle_optimizer() to make sure param_requires_grad_state is properly reset.

Parameters

• optimizer – The optimizer to toggle.

• optimizer_idx – The index of the optimizer to toggle.

train(mode: bool = True) → T

Sets the module in training mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

Parameters

mode (bool) – whether to set training mode (True) or evaluation mode (False). Default: True.

Returns

self

Return type

Module

train_dataloader() → Union[DataLoader, Sequence[DataLoader], Sequence[Sequence[DataLoader]], Sequence[Dict[str, DataLoader]], Dict[str, DataLoader], Dict[str, Dict[str, DataLoader]], Dict[str, Sequence[DataLoader]]]

Implement one or more PyTorch DataLoaders for training.

Returns

A collection of torch.utils.data.DataLoader specifying training samples. In the case of multiple dataloaders, please see this section.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

For data processing use the following pattern:

• download in prepare_data()

• process and split in setup()

However, the above are only necessary for distributed processing.

Warning

do not assign state in prepare_data

• fit()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware. There is no need to set it yourself.

Example:

# single dataloader
def train_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=True, transform=transform,
                    download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=True
    )
    return loader

# multiple dataloaders, return as list
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a list of tensors: [batch_mnist, batch_cifar]
    return [mnist_loader, cifar_loader]

# multiple dataloader, return as dict
def train_dataloader(self):
    mnist = MNIST(...)
    cifar = CIFAR(...)
    mnist_loader = torch.utils.data.DataLoader(
        dataset=mnist, batch_size=self.batch_size, shuffle=True
    )
    cifar_loader = torch.utils.data.DataLoader(
        dataset=cifar, batch_size=self.batch_size, shuffle=True
    )
    # each batch will be a dict of tensors: {'mnist': batch_mnist, 'cifar': batch_cifar}
    return {'mnist': mnist_loader, 'cifar': cifar_loader}

training_epoch_end(outputs: List[Union[Tensor, Dict[str, Any]]]) → None

Called at the end of the training epoch with the outputs of all training steps. Use this in case you need to do something with all the outputs returned by training_step().

# the pseudocode for these calls
train_outs = []
for train_batch in train_data:
    out = training_step(train_batch)
    train_outs.append(out)
training_epoch_end(train_outs)

Parameters

outputs – List of outputs you defined in training_step(). If there are multiple optimizers or when using truncated_bptt_steps > 0, the lists have the dimensions (n_batches, tbptt_steps, n_optimizers). Dimensions of length 1 are squeezed.

Returns

None

Note

If this method is not overridden, this won’t be called.

def training_epoch_end(self, training_step_outputs):
    # do something with all training_step outputs
    for out in training_step_outputs:
        ...

training_step(batch, batch_idx)

Here you compute and return the training loss and some additional metrics for e.g. the progress bar or logger.

Parameters

• batch (Tensor | (Tensor, …) | [Tensor, …]) – The output of your DataLoader. A tensor, tuple or list.

• batch_idx (int) – Integer displaying index of this batch

• optimizer_idx (int) – When using multiple optimizers, this argument will also be present.

• hiddens (Any) – Passed in if :paramref:`~pytorch_lightning.core.module.LightningModule.truncated_bptt_steps` > 0.

Returns

Any of.

• Tensor - The loss tensor

• dict - A dictionary. Can include any keys, but must include the key 'loss'

• None - Training will skip to the next batch. This is only for automatic optimization.

  This is not supported for multi-GPU, TPU, IPU, or DeepSpeed.

In this step you’d normally do the forward pass and calculate the loss for a batch. You can also do fancier things like multiple forward passes or something model specific.

Example:

def training_step(self, batch, batch_idx):
    x, y, z = batch
    out = self.encoder(x)
    loss = self.loss(out, x)
    return loss

If you define multiple optimizers, this step will be called with an additional optimizer_idx parameter.

# Multiple optimizers (e.g.: GANs)
def training_step(self, batch, batch_idx, optimizer_idx):
    if optimizer_idx == 0:
        # do training_step with encoder
        ...
    if optimizer_idx == 1:
        # do training_step with decoder
        ...

If you add truncated back propagation through time you will also get an additional argument with the hidden states of the previous step.

# Truncated back-propagation through time
def training_step(self, batch, batch_idx, hiddens):
    # hiddens are the hidden states from the previous truncated backprop step
    out, hiddens = self.lstm(data, hiddens)
    loss = ...
    return {"loss": loss, "hiddens": hiddens}

Note

The loss value shown in the progress bar is smoothed (averaged) over the last values, so it differs from the actual loss returned in train/validation step.

Note

When accumulate_grad_batches > 1, the loss returned here will be automatically normalized by accumulate_grad_batches internally.

training_step_end(step_output: Union[Tensor, Dict[str, Any]]) → Union[Tensor, Dict[str, Any]]

Use this when training with dp because training_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [training_step(sub_batch) for sub_batch in sub_batches]
training_step_end(step_output)

Parameters

step_output – What you return in training_step for each batch part.

Returns

Anything

When using the DP strategy, only a portion of the batch is inside the training_step:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)

    # softmax uses only a portion of the batch in the denominator
    loss = self.softmax(out)
    loss = nce_loss(loss)
    return loss

If you wish to do something with all the parts of the batch, then use this method to do it:

def training_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    return {"pred": out}


def training_step_end(self, training_step_outputs):
    gpu_0_pred = training_step_outputs[0]["pred"]
    gpu_1_pred = training_step_outputs[1]["pred"]
    gpu_n_pred = training_step_outputs[n]["pred"]

    # this softmax now uses the full batch
    loss = nce_loss([gpu_0_pred, gpu_1_pred, gpu_n_pred])
    return loss

See also

See the Multi GPU Training guide for more details.

transfer_batch_to_device(batch: Any, device: device, dataloader_idx: int) → Any

Override this hook if your DataLoader returns tensors wrapped in a custom data structure.

The data types listed below (and any arbitrary nesting of them) are supported out of the box:

• torch.Tensor or anything that implements .to(…)

• list

• dict

• tuple

For anything else, you need to define how the data is moved to the target device (CPU, GPU, TPU, …).

Note

This hook should only transfer the data and not modify it, nor should it move the data to any other device than the one passed in as argument (unless you know what you are doing). To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters

• batch – A batch of data that needs to be transferred to a new device.

• device – The target device as defined in PyTorch.

• dataloader_idx – The index of the dataloader to which the batch belongs.

Returns

A reference to the data on the new device.

Example:

def transfer_batch_to_device(self, batch, device, dataloader_idx):
    if isinstance(batch, CustomBatch):
        # move all tensors in your custom data structure to the device
        batch.samples = batch.samples.to(device)
        batch.targets = batch.targets.to(device)
    elif dataloader_idx == 0:
        # skip device transfer for the first dataloader or anything you wish
        pass
    else:
        batch = super().transfer_batch_to_device(data, device, dataloader_idx)
    return batch

Raises

• MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

• MisconfigurationException – If using IPUs, Trainer(accelerator='ipu').

See also

• move_data_to_device()

• apply_to_collection()

transform(loader: DataLoader)

Parameters

loader – a dataloader that matches the structure of that used for training

Returns

transformed views

property truncated_bptt_steps: int

Enables Truncated Backpropagation Through Time in the Trainer when set to a positive integer.

It represents the number of times training_step() gets called before backpropagation. If this is > 0, the training_step() receives an additional argument hiddens and is expected to return a hidden state.

type(dst_type: Union[str, dtype]) → Self

See torch.nn.Module.type().

unfreeze() → None

Unfreeze all parameters for training.

model = MyLightningModule(...)
model.unfreeze()

untoggle_optimizer(optimizer_idx: int) → None

Resets the state of required gradients that were toggled with toggle_optimizer().

This is only called automatically when automatic optimization is enabled and multiple optimizers are used.

Parameters

optimizer_idx – The index of the optimizer to untoggle.

val_dataloader() → Union[DataLoader, Sequence[DataLoader]]

Implement one or multiple PyTorch DataLoaders for validation.

The dataloader you return will not be reloaded unless you set :paramref:`~pytorch_lightning.trainer.Trainer.reload_dataloaders_every_n_epochs` to a positive integer.

It’s recommended that all data downloads and preparation happen in prepare_data().

• fit()

• validate()

• prepare_data()

• setup()

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Returns

A torch.utils.data.DataLoader or a sequence of them specifying validation samples.

Examples:

def val_dataloader(self):
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (1.0,))])
    dataset = MNIST(root='/path/to/mnist/', train=False,
                    transform=transform, download=True)
    loader = torch.utils.data.DataLoader(
        dataset=dataset,
        batch_size=self.batch_size,
        shuffle=False
    )

    return loader

# can also return multiple dataloaders
def val_dataloader(self):
    return [loader_a, loader_b, ..., loader_n]

Note

If you don’t need a validation dataset and a validation_step(), you don’t need to implement this method.

Note

In the case where you return multiple validation dataloaders, the validation_step() will have an argument dataloader_idx which matches the order here.

validation_epoch_end(outputs: Union[List[Union[Tensor, Dict[str, Any]]], List[List[Union[Tensor, Dict[str, Any]]]]]) → None

Called at the end of the validation epoch with the outputs of all validation steps.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

outputs – List of outputs you defined in validation_step(), or if there are multiple dataloaders, a list containing a list of outputs for each dataloader.

Returns

None

Note

If you didn’t define a validation_step(), this won’t be called.

Examples

With a single dataloader:

def validation_epoch_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

With multiple dataloaders, outputs will be a list of lists. The outer list contains one entry per dataloader, while the inner list contains the individual outputs of each validation step for that dataloader.

def validation_epoch_end(self, outputs):
    for dataloader_output_result in outputs:
        dataloader_outs = dataloader_output_result.dataloader_i_outputs

    self.log("final_metric", final_value)

validation_step(batch, batch_idx)

Operates on a single batch of data from the validation set. In this step you’d might generate examples or calculate anything of interest like accuracy.

# the pseudocode for these calls
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    val_outs.append(out)
validation_epoch_end(val_outs)

Parameters

• batch – The output of your DataLoader.

• batch_idx – The index of this batch.

• dataloader_idx – The index of the dataloader that produced this batch. (only if multiple val dataloaders used)

Returns

• Any object or value

• None - Validation will skip to the next batch

# pseudocode of order
val_outs = []
for val_batch in val_data:
    out = validation_step(val_batch)
    if defined("validation_step_end"):
        out = validation_step_end(out)
    val_outs.append(out)
val_outs = validation_epoch_end(val_outs)

# if you have one val dataloader:
def validation_step(self, batch, batch_idx):
    ...


# if you have multiple val dataloaders:
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    ...

Examples:

# CASE 1: A single validation dataset
def validation_step(self, batch, batch_idx):
    x, y = batch

    # implement your own
    out = self(x)
    loss = self.loss(out, y)

    # log 6 example images
    # or generated text... or whatever
    sample_imgs = x[:6]
    grid = torchvision.utils.make_grid(sample_imgs)
    self.logger.experiment.add_image('example_images', grid, 0)

    # calculate acc
    labels_hat = torch.argmax(out, dim=1)
    val_acc = torch.sum(y == labels_hat).item() / (len(y) * 1.0)

    # log the outputs!
    self.log_dict({'val_loss': loss, 'val_acc': val_acc})

If you pass in multiple val dataloaders, validation_step() will have an additional argument. We recommend setting the default value of 0 so that you can quickly switch between single and multiple dataloaders.

# CASE 2: multiple validation dataloaders
def validation_step(self, batch, batch_idx, dataloader_idx=0):
    # dataloader_idx tells you which dataset this is.
    ...

Note

If you don’t need to validate you don’t need to implement this method.

Note

When the validation_step() is called, the model has been put in eval mode and PyTorch gradients have been disabled. At the end of validation, the model goes back to training mode and gradients are enabled.

validation_step_end(*args: Any, **kwargs: Any) → Optional[Union[Tensor, Dict[str, Any]]]

Use this when validating with dp because validation_step() will operate on only part of the batch. However, this is still optional and only needed for things like softmax or NCE loss.

Note

If you later switch to ddp or some other mode, this will still be called so that you don’t have to change your code.

# pseudocode
sub_batches = split_batches_for_dp(batch)
step_output = [validation_step(sub_batch) for sub_batch in sub_batches]
validation_step_end(step_output)

Parameters

step_output – What you return in validation_step() for each batch part.

Returns

None or anything

# WITHOUT validation_step_end
# if used in DP, this batch is 1/num_gpus large
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self.encoder(x)
    loss = self.softmax(out)
    loss = nce_loss(loss)
    self.log("val_loss", loss)


# --------------
# with validation_step_end to do softmax over the full batch
def validation_step(self, batch, batch_idx):
    # batch is 1/num_gpus big
    x, y = batch

    out = self(x)
    return out


def validation_step_end(self, val_step_outputs):
    for out in val_step_outputs:
        ...

See also

See the Multi GPU Training guide for more details.

xpu(device: Optional[Union[device, int]] = None) → T

Moves all model parameters and buffers to the XPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on XPU while being optimized.

Note

This method modifies the module in-place.

Parameters

device (int, optional) – if specified, all parameters will be copied to that device

Returns

self

Return type

Module

zero_grad(set_to_none: bool = False) → None

Sets gradients of all model parameters to zero. See similar function under torch.optim.Optimizer for more context.

Parameters

set_to_none (bool) – instead of setting to zero, set the grads to None. See torch.optim.Optimizer.zero_grad() for details.

Objectives

class cca_zoo.deepmodels.objectives.MCCA(latent_dims: int, r: float = 0, eps: float = 0.001)

Bases: object

Differentiable MCCA Loss. Loss() method takes the outputs of each view’s network and solves the multiset eigenvalue problem as in e.g. https://arxiv.org/pdf/2005.11914.pdf

Parameters

• latent_dims – the number of latent dimensions

• r – regularisation as in regularized CCA. Makes the problem well posed when batch size is similar to

the number of latent dimensions :param eps: an epsilon parameter used in some operations

class cca_zoo.deepmodels.objectives.GCCA(latent_dims: int, r: float = 0, eps: float = 0.001)

Bases: object

Differentiable GCCA Loss. Loss() method takes the outputs of each view’s network and solves the generalized CCA eigenproblem as in https://arxiv.org/pdf/2005.11914.pdf

Parameters

• latent_dims – the number of latent dimensions

• r – regularisation as in regularized CCA. Makes the problem well posed when batch size is similar to

the number of latent dimensions :param eps: an epsilon parameter used in some operations

class cca_zoo.deepmodels.objectives.CCA(latent_dims: int, r: float = 0, eps: float = 0.001)

Bases: object

Differentiable CCA Loss. Loss() method takes the outputs of each view’s network and solves the CCA problem as in Andrew’s original paper

Parameters

• latent_dims – the number of latent dimensions

• r – regularisation as in regularized CCA. Makes the problem well posed when batch size is similar to the number of latent dimensions

• eps – an epsilon parameter used in some operations

class cca_zoo.deepmodels.objectives.TCCA(latent_dims: int, r: float = 0, eps: float = 0.0001)

Bases: object

Differentiable TCCA Loss.

Parameters

• latent_dims – the number of latent dimensions

• r – regularisation as in regularized CCA. Makes the problem well posed when batch size is similar to the number of latent dimensions

• eps – an epsilon parameter used in some operations

Callbacks

class cca_zoo.deepmodels.callbacks.CorrelationCallback

Bases: Callback

on_validation_epoch_end(trainer: Trainer, pl_module: LightningModule) → None

Called when the val epoch ends.

class cca_zoo.deepmodels.callbacks.GenerativeCallback

Bases: Callback

on_validation_epoch_end(trainer: Trainer, pl_module: LightningModule) → None

Called when the val epoch ends.

Architectures

class cca_zoo.deepmodels.architectures.Encoder(latent_dims: int, variational: bool = False, feature_size: int = 784, layer_sizes: Optional[tuple] = None, activation=LeakyReLU(negative_slope=0.01), dropout=0)

Bases: _BaseEncoder

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(x)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class cca_zoo.deepmodels.architectures.Decoder(latent_dims: int, feature_size: int = 784, layer_sizes: Optional[tuple] = None, activation=LeakyReLU(negative_slope=0.01), dropout=0)

Bases: _BaseDecoder

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(x)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class cca_zoo.deepmodels.architectures.CNNEncoder(latent_dims: int, variational: bool = False, feature_size: Iterable = (28, 28), channels: Optional[tuple] = None, kernel_sizes: Optional[tuple] = None, stride: Optional[tuple] = None, padding: Optional[tuple] = None, activation=LeakyReLU(negative_slope=0.01), dropout=0)

Bases: _BaseEncoder

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(x)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class cca_zoo.deepmodels.architectures.CNNDecoder(latent_dims: int, feature_size: Iterable = (28, 28), channels: Optional[tuple] = None, kernel_sizes=None, strides=None, paddings=None, activation=LeakyReLU(negative_slope=0.01), dropout=0)

Bases: _BaseDecoder

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(x)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class cca_zoo.deepmodels.architectures.LinearEncoder(latent_dims: int, feature_size: int, variational: bool = False)

Bases: _BaseEncoder

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(x)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class cca_zoo.deepmodels.architectures.LinearDecoder(latent_dims: int, feature_size: int)

Bases: _BaseDecoder

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(x)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.