deeptab.data

class deeptab.data.TabularDataset(cat_features_list, num_features_list, embeddings_list=None, labels=None, return_batch_object=False)[source]

Custom dataset for handling structured tabular data with separate categorical and numerical features.

This dataset is task-agnostic and simply stores and retrieves features and labels without any task-specific preprocessing. Label dtype conversion should be handled externally by the DataModule or training logic.

Parameters:

cat_features_list (list of Tensors) – A list of tensors representing the categorical features.
num_features_list (list of Tensors) – A list of tensors representing the numerical features.
embeddings_list (list of Tensors, optional) – A list of tensors representing the embeddings.
labels (Tensor, optional) – A tensor of labels. If None, the dataset is used for prediction.
return_batch_object (bool, default=False) – If True, returns a TabularBatch object instead of a tuple. For backward compatibility, defaults to False.

class deeptab.data.TabularDataModule(preprocessor, batch_size, shuffle, regression, X_val=None, y_val=None, val_size=0.2, random_state=101, stratify=True, sampler=None, **dataloader_kwargs)[source]

A PyTorch Lightning data module for managing training and validation data loaders in a structured way.

This class simplifies the process of batch-wise data loading for training and validation datasets during the training loop, and is particularly useful when working with PyTorch Lightning’s training framework.

Parameters:

preprocessor – object An instance of your preprocessor class.
batch_size – int Size of batches for the DataLoader.
shuffle – bool Whether to shuffle the training data in the DataLoader.
X_val – DataFrame or None, optional Validation features. If None, uses train-test split.
y_val – array-like or None, optional Validation labels. If None, uses train-test split.
val_size – float, optional Proportion of data to include in the validation split if X_val and y_val are None.
random_state – int, optional Random seed for reproducibility in data splitting.
regression – bool, optional Whether the problem is regression (True) or classification (False).
stratify – bool, optional Whether to stratify the validation split on the labels for classification tasks. Ignored for regression. Defaults to True.

classmethod from_datasets(train_dataset=None, val_dataset=None, test_dataset=None, predict_dataset=None, batch_size=1, num_workers=0, **datamodule_kwargs)

Create an instance from torch.utils.data.Dataset.

Parameters:

train_dataset (Union[Dataset, Iterable[Dataset], None]) – Optional dataset or iterable of datasets to be used for train_dataloader()
val_dataset (Union[Dataset, Iterable[Dataset], None]) – Optional dataset or iterable of datasets to be used for val_dataloader()
test_dataset (Union[Dataset, Iterable[Dataset], None]) – Optional dataset or iterable of datasets to be used for test_dataloader()
predict_dataset (Union[Dataset, Iterable[Dataset], None]) – Optional dataset or iterable of datasets to be used for predict_dataloader()
batch_size (int) – Batch size to use for each dataloader. Default is 1. This parameter gets forwarded to the __init__ if the datamodule has such a name defined in its signature.
num_workers (int) – Number of subprocesses to use for data loading. 0 means that the data will be loaded in the main process. Number of CPUs available. This parameter gets forwarded to the __init__ if the datamodule has such a name defined in its signature.
**datamodule_kwargs (Any) – Additional parameters that get passed down to the datamodule’s __init__.

Return type:

LightningDataModule

property hparams: 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

load_from_checkpoint(cls, checkpoint_path, map_location=None, hparams_file=None, weights_only=None, **kwargs)

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

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

Parameters:

checkpoint_path (Union[str, Path, IO]) – Path to checkpoint. This can also be a URL, or file-like object
map_location (Union[device, str, int, Callable[[UntypedStorage, str], Optional[UntypedStorage]], dict[Union[device, str, int], Union[device, str, int]], None]) – 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 (Union[str, Path, None]) –
Optional path to a .yaml or .csv file with hierarchical structure as in this example:
```
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 LightningDataModule for use.

If your datamodule’s hparams argument is Namespace and .yaml file has hierarchical structure, you need to refactor your datamodule to treat hparams as dict.
weights_only (Optional[bool]) – If True, restricts loading to state_dicts of plain torch.Tensor and other primitive types. If loading a checkpoint from a trusted source that contains an nn.Module, use weights_only=False. If loading checkpoint from an untrusted source, we recommend using weights_only=True. For more information, please refer to the PyTorch Developer Notes on Serialization Semantics.
**kwargs (Any) – Any extra keyword args needed to init the datamodule. Can also be used to override saved hyperparameter values.

Return type:

Self

Returns:

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

Note

load_from_checkpoint is a class method. You must use your LightningDataModule class to call it instead of the LightningDataModule instance, or a TypeError will be raised.

Example:

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

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

# override some of the params with new values
datamodule = MyLightningDataModule.load_from_checkpoint(
    PATH,
    batch_size=32,
    num_workers=10,
)

load_state_dict(state_dict)

Called when loading a checkpoint, implement to reload datamodule state given datamodule state_dict.

Parameters:: state_dict (dict[str, Any]) – the datamodule state returned by state_dict.
Return type:: None

on_after_batch_transfer(batch, dataloader_idx)

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.

Parameters:

batch (Any) – A batch of data that needs to be altered or augmented.
dataloader_idx (int) – The index of the dataloader to which the batch belongs.

Return type:

Any

Returns:

A batch of data

Example:

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

See also

on_after_batch_transfer()
transfer_batch_to_device()

on_exception(exception)

Called when the trainer execution is interrupted by an exception.

Return type:: None

preprocess_data(X_train, y_train, X_val=None, y_val=None, embeddings_train=None, embeddings_val=None, val_size=0.2, random_state=101)[source]

Preprocesses the training and validation data.

Parameters:

X_train (DataFrame or array-like, shape (n_samples_train, n_features)) – Training feature set.
y_train (array-like, shape (n_samples_train,)) – Training target values.
embeddings_train (array-like or list of array-like, optional) – Training embeddings if available.
X_val (DataFrame or array-like, shape (n_samples_val, n_features), optional) – Validation feature set. If None, a validation set will be created from X_train.
y_val (array-like, shape (n_samples_val,), optional) – Validation target values. If None, a validation set will be created from y_train.
embeddings_val (array-like or list of array-like, optional) – Validation embeddings if available.
val_size (float, optional) – Proportion of data to include in the validation split if X_val and y_val are None.
random_state (int, optional) – Random seed for reproducibility in data splitting.

Return type:

None

remove_ignored_hparams(ignore_list)

Remove ignored hyperparameters from the stored state.

This allows derived classes to drop hyperparameters previously saved by base classes.

Parameters:: ignore_list (list[str]) – Names of hyperparameters to remove.
Return type:: None

save_hyperparameters(*args, ignore=None, frame=None, logger=True)

Save arguments to hparams attribute.

Parameters:

args (Any) – single object of dict, NameSpace or OmegaConf or string names or arguments from class __init__
ignore (Union[Sequence[str], str, None]) – an argument name or a list of argument names from class __init__ to be ignored
frame (Optional[FrameType]) – a frame object. Default is None
logger (bool) – Whether to send the hyperparameters to the logger. Default: True

Return type:

None

Example::

>>> from lightning.pytorch.core.mixins import HyperparametersMixin
>>> 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

>>> from lightning.pytorch.core.mixins import HyperparametersMixin
>>> 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

>>> from lightning.pytorch.core.mixins import HyperparametersMixin
>>> 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

>>> from lightning.pytorch.core.mixins import HyperparametersMixin
>>> 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

property schema: FeatureSchema | None

Get the feature schema after preprocessing.

Returns:: Feature schema with metadata about categorical, numerical, and embedding features, or None if preprocessing hasn’t been done yet.
Return type:: FeatureSchema or None

setup(stage)[source]

Transform the data and create DataLoaders.

state_dict()

Called when saving a checkpoint, implement to generate and save datamodule state.

Return type:: dict[str, Any]
Returns:: A dictionary containing datamodule state.

teardown(stage)

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

Parameters:: stage (str) – either 'fit', 'validate', 'test', or 'predict'
Return type:: None

test_dataloader()[source]

Returns the test dataloader.

Returns:: DataLoader instance for the test dataset.
Return type:: DataLoader

train_dataloader()[source]

Returns the training dataloader.

Returns:: DataLoader instance for the training dataset.
Return type:: DataLoader

transfer_batch_to_device(batch, device, dataloader_idx)

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.

Parameters:

batch (Any) – A batch of data that needs to be transferred to a new device.
device (device) – The target device as defined in PyTorch.
dataloader_idx (int) – The index of the dataloader to which the batch belongs.

Return type:

Any

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(batch, device, dataloader_idx)
    return batch