vision_unlearning.datasets.base
===============================

.. py:module:: vision_unlearning.datasets.base


Attributes
----------

.. autoapisummary::

   vision_unlearning.datasets.base.logger


Exceptions
----------

.. autoapisummary::

   vision_unlearning.datasets.base.SplitNotAvailableError


Classes
-------

.. autoapisummary::

   vision_unlearning.datasets.base.UnlearnDatasetSplit
   vision_unlearning.datasets.base.UnlearnDatasetSplitMode
   vision_unlearning.datasets.base.UnlearnDataset


Module Contents
---------------

.. py:data:: logger

.. py:class:: UnlearnDatasetSplit

   Bases: :py:obj:`enum.Enum`


   Generic enumeration.

   Derive from this class to define new enumerations.


   .. py:attribute:: Train
      :value: 'train'



   .. py:attribute:: Validation
      :value: 'validation'



   .. py:attribute:: Test
      :value: 'test'



   .. py:attribute:: Train_retain
      :value: 'train_retain'



   .. py:attribute:: Train_retain_MIA
      :value: 'train_retain_mia'



   .. py:attribute:: Train_forget
      :value: 'train_forget'



   .. py:attribute:: Test_retain
      :value: 'test_retain'



   .. py:attribute:: Test_forget
      :value: 'test_forget'



   .. py:attribute:: Validation_retain
      :value: 'validation_retain'



   .. py:attribute:: Validation_forget
      :value: 'validation_forget'



.. py:class:: UnlearnDatasetSplitMode

   Bases: :py:obj:`enum.Enum`


   Generic enumeration.

   Derive from this class to define new enumerations.


   .. py:attribute:: Class
      :value: 'class'



   .. py:attribute:: Random
      :value: 'random'



   .. py:attribute:: Temporal
      :value: 'temporal'



.. py:exception:: SplitNotAvailableError

   Bases: :py:obj:`Exception`


   Common base class for all non-exit exceptions.


.. py:class:: UnlearnDataset(/, **data: Any)

   Bases: :py:obj:`pydantic.BaseModel`, :py:obj:`abc.ABC`


   Wrapper around huggingface datasets
   Organize the forget-retain splits


   .. py:attribute:: model_config

      Configuration for the model, should be a dictionary conforming to [`ConfigDict`][pydantic.config.ConfigDict].


   .. py:attribute:: split_mode
      :type:  UnlearnDatasetSplitMode


   .. py:attribute:: split_kwargs
      :type:  dict


   .. py:attribute:: _dataset_splits
      :type:  Dict[UnlearnDatasetSplit, Union[torch.utils.data.Subset, torchvision.datasets.vision.VisionDataset]]


   .. py:attribute:: _classes
      :type:  Optional[List[str]]
      :value: None



   .. py:attribute:: _n_classes
      :type:  int
      :value: 0



   .. py:attribute:: mean
      :type:  Optional[Sequence[float]]
      :value: None



   .. py:attribute:: std
      :type:  Optional[Sequence[float]]
      :value: None



   .. py:method:: model_post_init(__context: Optional[dict]) -> None

      Override this method to perform additional initialization after `__init__` and `model_construct`.
      This is useful if you want to do some validation that requires the entire model to be initialized.



   .. py:method:: _load() -> None
      :abstractmethod:


      Load the dataset from disk or download it.
      Side effects: updates the properties _dataset_splits, _classes, _n_classes



   .. py:method:: _split() -> None

      Split the dataset based on the specified mode.
      Side effects: updates the property dataset_splits
      Raised exceptions: none



   .. py:method:: _split_class(forget: List[str | int] | str | int) -> None


   .. py:method:: _split_random(n_forget: int, seed: int = 42) -> None
      :abstractmethod:



   .. py:method:: _split_temporal(n_forget: int) -> None
      :abstractmethod:



   .. py:method:: get_loader(split: UnlearnDatasetSplit, batchsize: int, shuffle: bool = True, num_workers: int = 0, pin_memory: bool = True) -> Optional[torch.utils.data.DataLoader]

      Return this split for this dataset.
      Side effects: none
      Raised exceptions: SplitNotAvailableError, if the requested split is not available



   .. py:method:: get_splits() -> Dict[UnlearnDatasetSplit, Union[torch.utils.data.Subset, torchvision.datasets.vision.VisionDataset]]

      Return the available splits.
      Side effects: none
      Raised exceptions: none



   .. py:method:: denormalize(normalized: torch.Tensor) -> torch.Tensor


   .. py:method:: save(path: str, format: Literal['pkl', 'jpg'] = 'pkl', save_unsplit: bool = False) -> None

      Save each split to disk.
      Side effects: saves files to disk
      Raised exceptions: OS-related errors



   .. py:method:: make_prompt_for_label(label: int) -> str