======= Dataset ======= This module contains classes and functions related to data handling. ValidSplit ---------- This class is responsible for performing the :class:`.NeuralNet`\'s internal cross validation. For this, it sticks closely to the sklearn standards. For more information on how sklearn handles cross validation, look `here `_. The first argument that :class:`.ValidSplit` takes is ``cv``. It works analogously to the ``cv`` argument from sklearn :class:`~sklearn.model_selection.GridSearchCV`, :func:`~sklearn.model_selection.cross_val_score`, etc. For those not familiar, here is a short explanation of what you may pass: - ``None``: Use the default 3-fold cross validation. - integer: Specifies the number of folds in a ``(Stratified)KFold``, - float: Represents the proportion of the dataset to include in the validation split (e.g. ``0.2`` for 20%). - An object to be used as a cross-validation generator. - An iterable yielding train, validation splits. Furthermore, :class:`.ValidSplit` takes a ``stratified`` argument that determines whether a stratified split should be made (only makes sense for discrete targets), and a ``random_state`` argument, which is used in case the cross validation split has a random component. One difference to sklearn\'s cross validation is that skorch makes only a single split. In sklearn, you would expect that in a 5-fold cross validation, the model is trained 5 times on the different combination of folds. This is often not desirable for neural networks, since training takes a lot of time. Therefore, skorch only ever makes one split. If you would like to have all splits, you can still use skorch in conjunction with the sklearn functions, as you would do with any other sklearn\-compatible estimator. Just remember to set ``train_split=None``, so that the whole dataset is used for training. Below is shown an example of making out-of-fold predictions with skorch and sklearn: .. code:: python net = NeuralNetClassifier( module=MyModule, train_split=None, ) from sklearn.model_selection import cross_val_predict y_pred = cross_val_predict(net, X, y, cv=5) Dataset ------- In PyTorch, we have the concept of a :class:`~torch.utils.data.Dataset` and a :class:`~torch.utils.data.DataLoader`. The former is purely the container of the data and only needs to implement ``__len__()`` and ``__getitem__()``. The latter does the heavy lifting, such as sampling, shuffling, and distributed processing. skorch uses the PyTorch :class:`~torch.utils.data.DataLoader`\s by default. skorch supports PyTorch's :class:`~torch.utils.data.Dataset` when calling :func:`~skorch.net.NeuralNet.fit` or :func:`~skorch.net.NeuralNet.partial_fit`. Details on how to use PyTorch's :class:`~torch.utils.data.Dataset` with skorch, can be found in :ref:`faq_how_do_i_use_a_pytorch_dataset_with_skorch`. In order to support other data formats, we provide our own :class:`.Dataset` class that is compatible with: - :class:`numpy.ndarray`\s - PyTorch :class:`~torch.Tensor`\s - scipy sparse CSR matrices - pandas DataFrames or Series Note that currently, sparse matrices are cast to dense arrays during batching, given that PyTorch support for sparse matrices is still very incomplete. If you would like to prevent that, you need to override the ``transform`` method of :class:`~torch.utils.data.Dataset`. In addition to the types above, you can pass dictionaries or lists of one of those data types, e.g. a dictionary of :class:`numpy.ndarray`\s. When you pass dictionaries, the keys of the dictionaries are used as the argument name for the :meth:`~torch.nn.Module.forward` method of the net's ``module``. Similarly, the column names of pandas ``DataFrame``\s are used as argument names. The example below should illustrate how to use this feature: .. code:: python import numpy as np import torch import torch.nn.functional as F class MyModule(torch.nn.Module): def __init__(self): super().__init__() self.dense_a = torch.nn.Linear(10, 100) self.dense_b = torch.nn.Linear(20, 100) self.output = torch.nn.Linear(200, 2) def forward(self, key_a, key_b): hid_a = F.relu(self.dense_a(key_a)) hid_b = F.relu(self.dense_b(key_b)) concat = torch.cat((hid_a, hid_b), dim=1) out = F.softmax(self.output(concat)) return out net = NeuralNetClassifier(MyModule) X = { 'key_a': np.random.random((1000, 10)).astype(np.float32), 'key_b': np.random.random((1000, 20)).astype(np.float32), } y = np.random.randint(0, 2, size=1000) net.fit(X, y) Note that the keys in the dictionary ``X`` exactly match the argument names in the :meth:`~torch.nn.Module.forward` method. This way, you can easily work with several different types of input features. The :class:`.Dataset` from skorch makes the assumption that you always have an ``X`` and a ``y``, where ``X`` represents the input data and ``y`` the target. However, you may leave ``y=None``, in which case :class:`.Dataset` returns a dummy variable. :class:`.Dataset` applies a transform final transform on the data before passing it on to the PyTorch :class:`~torch.utils.data.DataLoader`. By default, it replaces ``y`` by a dummy variable in case it is ``None``. If you would like to apply your own transformation on the data, you should subclass :class:`.Dataset` and override the :func:`~skorch.dataset.Dataset.transform` method, then pass your custom class to :class:`.NeuralNet` as the ``dataset`` argument. Preventing dictionary unpacking ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ As noted, when ``X`` is a dictionary, it is automatically unpacked when passed to the module's :meth:`~torch.nn.Module.forward` method. Sometimes, you may want to prevent this, e.g. because you're using a ``module`` from another library that expects a dict as input, or because the exact dict keys are unknown. This can be achieved by wrapping the original ``module`` and packing the dict again: .. code:: python from other_lib import ModuleExpectingDict class WrappedModule(ModuleExpectingDict): def forward(self, **kwargs): # catch **kwargs, pass as a dict return super().forward(kwargs) Similarly, wrapping the ``module`` can be used to make any other desired changes to the input arguments.