braindecode.classifier.EEGClassifier

class braindecode.classifier.EEGClassifier(*args, cropped=False, callbacks=None, iterator_train__shuffle=True, **kwargs)

Classifier that does not assume softmax activation. Calls loss function directly without applying log or anything.

Parameters
module: torch module (class or instance)

A PyTorch Module. In general, the uninstantiated class should be passed, although instantiated modules will also work.

criterion: torch criterion (class, default=torch.nn.NLLLoss)

Negative log likelihood loss. Note that the module should return probabilities, the log is applied during get_loss.

classes: None or list (default=None)

If None, the classes_ attribute will be inferred from the y data passed to fit. If a non-empty list is passed, that list will be returned as classes_. If the initial skorch behavior should be restored, i.e. raising an AttributeError, pass an empty list.

optimizer: torch optim (class, default=torch.optim.SGD)

The uninitialized optimizer (update rule) used to optimize the module

lr: float (default=0.01)

Learning rate passed to the optimizer. You may use lr instead of using optimizer__lr, which would result in the same outcome.

max_epochs: int (default=10)

The number of epochs to train for each fit call. Note that you may keyboard-interrupt training at any time.

batch_size: int (default=128)

Mini-batch size. Use this instead of setting iterator_train__batch_size and iterator_test__batch_size, which would result in the same outcome. If batch_size is -1, a single batch with all the data will be used during training and validation.

iterator_train: torch DataLoader

The default PyTorch DataLoader used for training data.

iterator_valid: torch DataLoader

The default PyTorch DataLoader used for validation and test data, i.e. during inference.

dataset: torch Dataset (default=skorch.dataset.Dataset)

The dataset is necessary for the incoming data to work with pytorch’s DataLoader. It has to implement the __len__ and __getitem__ methods. The provided dataset should be capable of dealing with a lot of data types out of the box, so only change this if your data is not supported. You should generally pass the uninitialized Dataset class and define additional arguments to X and y by prefixing them with dataset__. It is also possible to pass an initialzed Dataset, in which case no additional arguments may be passed.

train_split: None or callable (default=skorch.dataset.CVSplit(5))

If None, there is no train/validation split. Else, train_split should be a function or callable that is called with X and y data and should return the tuple dataset_train, dataset_valid. The validation data may be None.

predict_nonlinearity: callable, None, or ‘auto’ (default=’auto’)

The nonlinearity to be applied to the prediction. When set to ‘auto’, infers the correct nonlinearity based on the criterion (softmax for CrossEntropyLoss and sigmoid for BCEWithLogitsLoss). If it cannot be inferred or if the parameter is None, just use the identity function. Don’t pass a lambda function if you want the net to be pickleable.

In case a callable is passed, it should accept the output of the module (the first output if there is more than one), which is a PyTorch tensor, and return the transformed PyTorch tensor.

This can be useful, e.g., when predict_proba() should return probabilities but a criterion is used that does not expect probabilities. In that case, the module can return whatever is required by the criterion and the predict_nonlinearity transforms this output into probabilities.

The nonlinearity is applied only when calling predict() or predict_proba() but not anywhere else – notably, the loss is unaffected by this nonlinearity.

warm_start: bool (default=False)

Whether each fit call should lead to a re-initialization of the module (cold start) or whether the module should be trained further (warm start).

verbose: int (default=1)

This parameter controls how much print output is generated by the net and its callbacks. By setting this value to 0, e.g. the summary scores at the end of each epoch are no longer printed. This can be useful when running a hyperparameter search. The summary scores are always logged in the history attribute, regardless of the verbose setting.

device: str, torch.device (default=’cpu’)

The compute device to be used. If set to ‘cuda’, data in torch tensors will be pushed to cuda tensors before being sent to the module. If set to None, then all compute devices will be left unmodified.

cropped: bool (default=False)

Defines whether torch model passed to this class is cropped or not. Currently used for callbacks definition.

callbacks: None or list of strings or list of Callback instances (default=None)

More callbacks, in addition to those returned by get_default_callbacks. Each callback should inherit from skorch.callbacks.Callback. If not None, callbacks can be a list of strings specifying sklearn scoring functions (for scoring functions names see: https://scikit-learn.org/stable/modules/model_evaluation.html#scoring-parameter) or a list of callbacks where the callback names are inferred from the class name. Name conflicts are resolved by appending a count suffix starting with 1, e.g. EpochScoring_1. Alternatively, a tuple (name, callback) can be passed, where name should be unique. Callbacks may or may not be instantiated. The callback name can be used to set parameters on specific callbacks (e.g., for the callback with name 'print_log', use net.set_params(callbacks__print_log__keys_ignored=['epoch', 'train_loss'])).

iterator_train__shuffle: bool (default=True)

Defines whether train dataset will be shuffled. As skorch does not shuffle the train dataset by default this one overwrites this option.

Attributes
prefixes_: list of str

Contains the prefixes to special parameters. E.g., since there is the 'module' prefix, it is possible to set parameters like so: NeuralNet(..., optimizer__momentum=0.95).

cuda_dependent_attributes_: list of str

Contains a list of all attribute prefixes whose values depend on a CUDA device. If a NeuralNet trained with a CUDA-enabled device is unpickled on a machine without CUDA or with CUDA disabled, the listed attributes are mapped to CPU. Expand this list if you want to add other cuda-dependent attributes.

initialized_: bool

Whether the NeuralNet was initialized.

module_: torch module (instance)

The instantiated module.

criterion_: torch criterion (instance)

The instantiated criterion.

callbacks_: list of tuples

The complete (i.e. default and other), initialized callbacks, in a tuple with unique names.

classes_: array, shape (n_classes, )

A list of class labels known to the classifier.

Methods

check_data(X, y)

check_is_fitted([attributes])

Checks whether the net is initialized

evaluation_step(Xi[, training])

Perform a forward step to produce the output used for prediction and scoring.

fit(X, y, **fit_params)

See NeuralNet.fit.

fit_loop(X[, y, epochs])

The proper fit loop.

forward(X[, training, device])

Gather and concatenate the output from forward call with input data.

forward_iter(X[, training, device])

Yield outputs of module forward calls on each batch of data.

get_dataset(X[, y])

Get a dataset that contains the input data and is passed to the iterator.

get_iterator(dataset[, training, drop_index])

Get an iterator that allows to loop over the batches of the given data.

get_loss(y_pred, y_true, *args, **kwargs)

Return the loss for this batch by calling NeuralNet get_loss. Parameters ———- y_pred : torch tensor Predicted target values y_true : torch tensor True target values. X : input data, compatible with skorch.dataset.Dataset By default, you should be able to pass: * numpy arrays * torch tensors * pandas DataFrame or Series * scipy sparse CSR matrices * a dictionary of the former three * a list/tuple of the former three * a Dataset If this doesn’t work with your data, you have to pass a Dataset that can deal with the data. training : bool (default=False) Whether train mode should be used or not.

get_params_for(prefix)

Collect and return init parameters for an attribute.

get_params_for_optimizer(prefix, …)

Collect and return init parameters for an optimizer.

get_split_datasets(X[, y])

Get internal train and validation datasets.

get_train_step_accumulator()

Return the train step accumulator.

infer(x, **fit_params)

Perform a single inference step on a batch of data.

initialize()

Initializes all components of the NeuralNet and returns self.

initialize_callbacks()

Initializes all callbacks and save the result in the callbacks_ attribute.

initialize_criterion()

Initializes the criterion.

initialize_history()

Initializes the history.

initialize_module()

Initializes the module.

initialize_optimizer([triggered_directly])

Initialize the model optimizer.

load_params([f_params, f_optimizer, …])

Loads the the module’s parameters, history, and optimizer, not the whole object.

notify(method_name, **cb_kwargs)

Call the callback method specified in method_name with parameters specified in cb_kwargs.

on_batch_begin(net[, Xi, yi, training])

on_epoch_begin(net[, dataset_train, …])

on_epoch_end(net[, dataset_train, dataset_valid])

on_train_begin(net[, X, y])

on_train_end(net[, X, y])

partial_fit(X[, y, classes])

Fit the module.

predict(X)

Where applicable, return class labels for samples in X.

predict_proba(X)

Where applicable, return probability estimates for samples.

run_single_epoch(dataset, training, prefix, …)

Compute a single epoch of train or validation.

save_params([f_params, f_optimizer, …])

Saves the module’s parameters, history, and optimizer, not the whole object.

score(X, y[, sample_weight])

Return the mean accuracy on the given test data and labels.

set_params(**kwargs)

Set the parameters of this class.

train_step(Xi, yi, **fit_params)

Prepares a loss function callable and pass it to the optimizer, hence performing one optimization step.

train_step_single(Xi, yi, **fit_params)

Compute y_pred, loss value, and update net’s gradients.

validation_step(Xi, yi, **fit_params)

Perform a forward step using batched data and return the resulting loss.

get_default_callbacks

get_params

initialize_virtual_params

on_batch_end

on_grad_computed

predict_with_window_inds_and_ys

check_is_fitted(attributes=None, *args, **kwargs)

Checks whether the net is initialized

Parameters
attributesiterable of str or None (default=None)

All the attributes that are strictly required of a fitted net. By default, this is the module_ attribute.

Other arguments as in
``sklearn.utils.validation.check_is_fitted``.
Raises
skorch.exceptions.NotInitializedError

When the given attributes are not present.

evaluation_step(Xi, training=False)

Perform a forward step to produce the output used for prediction and scoring.

Therefore the module is set to evaluation mode by default beforehand which can be overridden to re-enable features like dropout by setting training=True.

fit(X, y, **fit_params)

See NeuralNet.fit.

In contrast to NeuralNet.fit, y is non-optional to avoid mistakenly forgetting about y. However, y can be set to None in case it is derived dynamically from X.

fit_loop(X, y=None, epochs=None, **fit_params)

The proper fit loop.

Contains the logic of what actually happens during the fit loop.

Parameters
Xinput data, compatible with skorch.dataset.Dataset

By default, you should be able to pass:

  • numpy arrays

  • torch tensors

  • pandas DataFrame or Series

  • scipy sparse CSR matrices

  • a dictionary of the former three

  • a list/tuple of the former three

  • a Dataset

If this doesn’t work with your data, you have to pass a Dataset that can deal with the data.

ytarget data, compatible with skorch.dataset.Dataset

The same data types as for X are supported. If your X is a Dataset that contains the target, y may be set to None.

epochsint or None (default=None)

If int, train for this number of epochs; if None, use self.max_epochs.

**fit_paramsdict

Additional parameters passed to the forward method of the module and to the self.train_split call.

forward(X, training=False, device='cpu')

Gather and concatenate the output from forward call with input data.

The outputs from self.module_.forward are gathered on the compute device specified by device and then concatenated using PyTorch cat(). If multiple outputs are returned by self.module_.forward, each one of them must be able to be concatenated this way.

Parameters
Xinput data, compatible with skorch.dataset.Dataset

By default, you should be able to pass:

  • numpy arrays

  • torch tensors

  • pandas DataFrame or Series

  • scipy sparse CSR matrices

  • a dictionary of the former three

  • a list/tuple of the former three

  • a Dataset

If this doesn’t work with your data, you have to pass a Dataset that can deal with the data.

trainingbool (default=False)

Whether to set the module to train mode or not.

devicestring (default=’cpu’)

The device to store each inference result on. This defaults to CPU memory since there is genereally more memory available there. For performance reasons this might be changed to a specific CUDA device, e.g. ‘cuda:0’.

Returns
y_infertorch tensor

The result from the forward step.

forward_iter(X, training=False, device='cpu')

Yield outputs of module forward calls on each batch of data. The storage device of the yielded tensors is determined by the device parameter.

Parameters
Xinput data, compatible with skorch.dataset.Dataset

By default, you should be able to pass:

  • numpy arrays

  • torch tensors

  • pandas DataFrame or Series

  • scipy sparse CSR matrices

  • a dictionary of the former three

  • a list/tuple of the former three

  • a Dataset

If this doesn’t work with your data, you have to pass a Dataset that can deal with the data.

trainingbool (default=False)

Whether to set the module to train mode or not.

devicestring (default=’cpu’)

The device to store each inference result on. This defaults to CPU memory since there is genereally more memory available there. For performance reasons this might be changed to a specific CUDA device, e.g. ‘cuda:0’.

Yields
yptorch tensor

Result from a forward call on an individual batch.

get_dataset(X, y=None)

Get a dataset that contains the input data and is passed to the iterator.

Override this if you want to initialize your dataset differently.

Parameters
Xinput data, compatible with skorch.dataset.Dataset

By default, you should be able to pass:

  • numpy arrays

  • torch tensors

  • pandas DataFrame or Series

  • scipy sparse CSR matrices

  • a dictionary of the former three

  • a list/tuple of the former three

  • a Dataset

If this doesn’t work with your data, you have to pass a Dataset that can deal with the data.

ytarget data, compatible with skorch.dataset.Dataset

The same data types as for X are supported. If your X is a Dataset that contains the target, y may be set to None.

Returns
dataset

The initialized dataset.

get_iterator(dataset, training=False, drop_index=True)

Get an iterator that allows to loop over the batches of the given data.

If self.iterator_train__batch_size and/or self.iterator_test__batch_size are not set, use self.batch_size instead.

Parameters
datasettorch Dataset (default=skorch.dataset.Dataset)

Usually, self.dataset, initialized with the corresponding data, is passed to get_iterator.

trainingbool (default=False)

Whether to use iterator_train or iterator_test.

Returns
iterator

An instantiated iterator that allows to loop over the mini-batches.

get_loss(y_pred, y_true, *args, **kwargs)

Return the loss for this batch by calling NeuralNet get_loss. Parameters ———- y_pred : torch tensor

Predicted target values

y_truetorch tensor

True target values.

Xinput data, compatible with skorch.dataset.Dataset
By default, you should be able to pass:
  • numpy arrays

  • torch tensors

  • pandas DataFrame or Series

  • scipy sparse CSR matrices

  • a dictionary of the former three

  • a list/tuple of the former three

  • a Dataset

If this doesn’t work with your data, you have to pass a Dataset that can deal with the data.

trainingbool (default=False)

Whether train mode should be used or not.

get_params_for(prefix)

Collect and return init parameters for an attribute.

Attributes could be, for instance, pytorch modules, criteria, or data loaders (for optimizers, use get_params_for_optimizer() instead). Use the returned arguments to initialize the given attribute like this:

# inside initialize_module method
kwargs = self.get_params_for('module')
self.module_ = self.module(**kwargs)

Proceed analogously for the criterion etc.

The reason to use this method is so that it’s possible to change the init parameters with set_params(), which in turn makes grid search and other similar things work.

Note that in general, as a user, you never have to deal with this method because initialize_module() etc. are already taking care of this. You only need to deal with this if you override initialize_module() (or similar methods) because you have some custom code that requires it.

Parameters
prefixstr

The name of the attribute whose arguments should be returned. E.g. for the module, it should be 'module'.

Returns
kwargsdict

Keyword arguments to be used as init parameters.

get_params_for_optimizer(prefix, named_parameters)

Collect and return init parameters for an optimizer.

Parse kwargs configuration for the optimizer identified by the given prefix. Supports param group assignment using wildcards:

optimizer__lr=0.05,
optimizer__param_groups=[
    ('rnn*.period', {'lr': 0.3, 'momentum': 0}),
    ('rnn0', {'lr': 0.1}),
]

Generally, use this method like this:

# inside initialize_optimizer method
named_params = self.module_.named_parameters()
pgroups, kwargs = self.get_params_for_optimizer('optimizer', named_params)
if 'lr' not in kwargs:
    kwargs['lr'] = self.lr
self.optimizer_ = self.optimizer(*pgroups, **kwargs)

The reason to use this method is so that it’s possible to change the init parameters with set_params(), which in turn makes grid search and other similar things work.

Note that in general, as a user, you never have to deal with this method because initialize_optimizer() is already taking care of this. You only need to deal with this if you override initialize_optimizer() because you have some custom code that requires it.

Parameters
prefixstr

The name of the optimizer whose arguments should be returned. Typically, this should just be 'optimizer'. There can be exceptions, however, e.g. if you want to use more than one optimizer.

named_parametersiterator

Iterator over the parameters of the module that is intended to be optimized. It’s the return value of my_module.named_parameters().

Returns
argstuple

All positional arguments for this optimizer (right now only one, the parameter groups).

kwargsdict

All other parameters for this optimizer, e.g. the learning rate.

get_split_datasets(X, y=None, **fit_params)

Get internal train and validation datasets.

The validation dataset can be None if self.train_split is set to None; then internal validation will be skipped.

Override this if you want to change how the net splits incoming data into train and validation part.

Parameters
Xinput data, compatible with skorch.dataset.Dataset

By default, you should be able to pass:

  • numpy arrays

  • torch tensors

  • pandas DataFrame or Series

  • scipy sparse CSR matrices

  • a dictionary of the former three

  • a list/tuple of the former three

  • a Dataset

If this doesn’t work with your data, you have to pass a Dataset that can deal with the data.

ytarget data, compatible with skorch.dataset.Dataset

The same data types as for X are supported. If your X is a Dataset that contains the target, y may be set to None.

**fit_paramsdict

Additional parameters passed to the self.train_split call.

Returns
dataset_train

The initialized training dataset.

dataset_valid

The initialized validation dataset or None

get_train_step_accumulator()

Return the train step accumulator.

By default, the accumulator stores and retrieves the first value from the optimizer call. Most optimizers make only one call, so first value is at the same time the only value.

In case of some optimizers, e.g. LBFGS, train_step_calc_gradient is called multiple times, as the loss function is evaluated multiple times per optimizer call. If you don’t want to return the first value in that case, override this method to return your custom accumulator.

infer(x, **fit_params)

Perform a single inference step on a batch of data.

Parameters
xinput data

A batch of the input data.

**fit_paramsdict

Additional parameters passed to the forward method of the module and to the self.train_split call.

initialize()

Initializes all components of the NeuralNet and returns self.

initialize_callbacks()

Initializes all callbacks and save the result in the callbacks_ attribute.

Both default_callbacks and callbacks are used (in that order). Callbacks may either be initialized or not, and if they don’t have a name, the name is inferred from the class name. The initialize method is called on all callbacks.

The final result will be a list of tuples, where each tuple consists of a name and an initialized callback. If names are not unique, a ValueError is raised.

initialize_criterion()

Initializes the criterion.

initialize_history()

Initializes the history.

initialize_module()

Initializes the module.

Note that if the module has learned parameters, those will be reset.

initialize_optimizer(triggered_directly=True)

Initialize the model optimizer. If self.optimizer__lr is not set, use self.lr instead.

Parameters
triggered_directlybool (default=True)

Only relevant when optimizer is re-initialized. Initialization of the optimizer can be triggered directly (e.g. when lr was changed) or indirectly (e.g. when the module was re-initialized). If and only if the former happens, the user should receive a message informing them about the parameters that caused the re-initialization.

load_params(f_params=None, f_optimizer=None, f_criterion=None, f_history=None, checkpoint=None, **kwargs)

Loads the the module’s parameters, history, and optimizer, not the whole object.

To save and load the whole object, use pickle.

f_params, f_optimizer, etc. uses PyTorch’s load().

If you’ve created a custom module, e.g. net.mymodule_, you can save that as well by passing f_mymodule.

Parameters
f_paramsfile-like object, str, None (default=None)

Path of module parameters. Pass None to not load.

f_optimizerfile-like object, str, None (default=None)

Path of optimizer. Pass None to not load.

f_criterionfile-like object, str, None (default=None)

Path of criterion. Pass None to not save

f_historyfile-like object, str, None (default=None)

Path to history. Pass None to not load.

checkpointCheckpoint, None (default=None)

Checkpoint to load params from. If a checkpoint and a f_* path is passed in, the f_* will be loaded. Pass None to not load.

Examples

>>> before = NeuralNetClassifier(mymodule)
>>> before.save_params(f_params='model.pkl',
>>>                    f_optimizer='optimizer.pkl',
>>>                    f_history='history.json')
>>> after = NeuralNetClassifier(mymodule).initialize()
>>> after.load_params(f_params='model.pkl',
>>>                   f_optimizer='optimizer.pkl',
>>>                   f_history='history.json')
notify(method_name, **cb_kwargs)

Call the callback method specified in method_name with parameters specified in cb_kwargs.

Method names can be one of: * on_train_begin * on_train_end * on_epoch_begin * on_epoch_end * on_batch_begin * on_batch_end

partial_fit(X, y=None, classes=None, **fit_params)

Fit the module.

If the module is initialized, it is not re-initialized, which means that this method should be used if you want to continue training a model (warm start).

Parameters
Xinput data, compatible with skorch.dataset.Dataset

By default, you should be able to pass:

  • numpy arrays

  • torch tensors

  • pandas DataFrame or Series

  • scipy sparse CSR matrices

  • a dictionary of the former three

  • a list/tuple of the former three

  • a Dataset

If this doesn’t work with your data, you have to pass a Dataset that can deal with the data.

ytarget data, compatible with skorch.dataset.Dataset

The same data types as for X are supported. If your X is a Dataset that contains the target, y may be set to None.

classesarray, sahpe (n_classes,)

Solely for sklearn compatibility, currently unused.

**fit_paramsdict

Additional parameters passed to the forward method of the module and to the self.train_split call.

predict(X)

Where applicable, return class labels for samples in X.

If the module’s forward method returns multiple outputs as a tuple, it is assumed that the first output contains the relevant information and the other values are ignored. If all values are relevant, consider using forward() instead.

Parameters
Xinput data, compatible with skorch.dataset.Dataset

By default, you should be able to pass:

  • numpy arrays

  • torch tensors

  • pandas DataFrame or Series

  • scipy sparse CSR matrices

  • a dictionary of the former three

  • a list/tuple of the former three

  • a Dataset

If this doesn’t work with your data, you have to pass a Dataset that can deal with the data.

Returns
y_prednumpy ndarray
predict_proba(X)

Where applicable, return probability estimates for samples.

If the module’s forward method returns multiple outputs as a tuple, it is assumed that the first output contains the relevant information and the other values are ignored. If all values are relevant, consider using forward() instead.

Parameters
Xinput data, compatible with skorch.dataset.Dataset

By default, you should be able to pass:

  • numpy arrays

  • torch tensors

  • pandas DataFrame or Series

  • scipy sparse CSR matrices

  • a dictionary of the former three

  • a list/tuple of the former three

  • a Dataset

If this doesn’t work with your data, you have to pass a Dataset that can deal with the data.

Returns
y_probanumpy ndarray
run_single_epoch(dataset, training, prefix, step_fn, **fit_params)

Compute a single epoch of train or validation.

Parameters
datasettorch Dataset

The initialized dataset to loop over.

trainingbool

Whether to set the module to train mode or not.

prefixstr

Prefix to use when saving to the history.

step_fncallable

Function to call for each batch.

**fit_paramsdict

Additional parameters passed to the step_fn.

save_params(f_params=None, f_optimizer=None, f_criterion=None, f_history=None, **kwargs)

Saves the module’s parameters, history, and optimizer, not the whole object.

To save the whole object, use pickle. This is necessary when you need additional learned attributes on the net, e.g. the classes_ attribute on skorch.classifier.NeuralNetClassifier.

f_params, f_optimizer, etc. use PyTorch’s save().

If you’ve created a custom module, e.g. net.mymodule_, you can save that as well by passing f_mymodule.

Parameters
f_paramsfile-like object, str, None (default=None)

Path of module parameters. Pass None to not save

f_optimizerfile-like object, str, None (default=None)

Path of optimizer. Pass None to not save

f_criterionfile-like object, str, None (default=None)

Path of criterion. Pass None to not save

f_historyfile-like object, str, None (default=None)

Path to history. Pass None to not save

Examples

>>> before = NeuralNetClassifier(mymodule)
>>> before.save_params(f_params='model.pkl',
...                    f_optimizer='optimizer.pkl',
...                    f_history='history.json')
>>> after = NeuralNetClassifier(mymodule).initialize()
>>> after.load_params(f_params='model.pkl',
...                   f_optimizer='optimizer.pkl',
...                   f_history='history.json')
score(X, y, sample_weight=None)

Return the mean accuracy on the given test data and labels.

In multi-label classification, this is the subset accuracy which is a harsh metric since you require for each sample that each label set be correctly predicted.

Parameters
Xarray-like of shape (n_samples, n_features)

Test samples.

yarray-like of shape (n_samples,) or (n_samples, n_outputs)

True labels for X.

sample_weightarray-like of shape (n_samples,), default=None

Sample weights.

Returns
scorefloat

Mean accuracy of self.predict(X) wrt. y.

set_params(**kwargs)

Set the parameters of this class.

Valid parameter keys can be listed with get_params().

Returns
self
train_step(Xi, yi, **fit_params)

Prepares a loss function callable and pass it to the optimizer, hence performing one optimization step.

Loss function callable as required by some optimizers (and accepted by all of them): https://pytorch.org/docs/master/optim.html#optimizer-step-closure

The module is set to be in train mode (e.g. dropout is applied).

Parameters
Xiinput data

A batch of the input data.

yitarget data

A batch of the target data.

**fit_paramsdict

Additional parameters passed to the forward method of the module and to the train_split call.

train_step_single(Xi, yi, **fit_params)

Compute y_pred, loss value, and update net’s gradients.

The module is set to be in train mode (e.g. dropout is applied).

Parameters
Xiinput data

A batch of the input data.

yitarget data

A batch of the target data.

**fit_paramsdict

Additional parameters passed to the forward method of the module and to the self.train_split call.

validation_step(Xi, yi, **fit_params)

Perform a forward step using batched data and return the resulting loss.

The module is set to be in evaluation mode (e.g. dropout is not applied).

Parameters
Xiinput data

A batch of the input data.

yitarget data

A batch of the target data.

**fit_paramsdict

Additional parameters passed to the forward method of the module and to the self.train_split call.