personalitylinmult.data.datamodule
==================================

.. py:module:: personalitylinmult.data.datamodule

.. autoapi-nested-parse::

   Lightning ``DataModule`` over the built FI HDF5 file.



Attributes
----------

.. autoapisummary::

   personalitylinmult.data.datamodule.logger


Classes
-------

.. autoapisummary::

   personalitylinmult.data.datamodule.FiDataModule


Functions
---------

.. autoapisummary::

   personalitylinmult.data.datamodule.seed_worker


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

.. py:data:: logger

   Module-level logger.

.. py:function:: seed_worker(worker_id: int) -> None

   Seed a dataloader worker's RNGs from torch's per-worker seed.

   Without this, every worker inherits the parent's numpy seed and augmentation
   draws repeat across workers.

   :param worker_id: Worker index, unused but required by the hook.
   :type worker_id: int


.. py:class:: FiDataModule(config: personalitylinmult.train.config.DataConfig, target_names: list[str])

   Bases: :py:obj:`lightning.LightningDataModule`


   Serves the FI splits from the single HDF5 artifact.

   :param config: Dataloader configuration.
   :type config: DataConfig
   :param target_names: Label columns for the active task, in head
                        order.
   :type target_names: list[str]

   .. attribute:: prepare_data_per_node

      If True, each LOCAL_RANK=0 will call prepare data.
      Otherwise only NODE_RANK=0, LOCAL_RANK=0 will prepare data.

   .. attribute:: allow_zero_length_dataloader_with_multiple_devices

      If True, dataloader with zero length within local rank is allowed.
      Default value is False.


   .. py:method:: setup(stage: str | None = None) -> None

      Construct the datasets needed for a stage.

      :param stage: ``"fit"``, ``"validate"``, ``"test"``, or
                    ``None`` for all.
      :type stage: str | None



   .. py:method:: train_dataloader() -> torch.utils.data.DataLoader

      Training loader.

      :returns: Shuffled, drop-last loader over the train split.
      :rtype: DataLoader

      :raises RuntimeError: If called before ``setup``.



   .. py:method:: val_dataloader() -> torch.utils.data.DataLoader

      Validation loader.

      :returns: Deterministic loader over the valid split.
      :rtype: DataLoader

      :raises RuntimeError: If called before ``setup``.



   .. py:method:: test_dataloader() -> torch.utils.data.DataLoader

      Test loader.

      :returns: Deterministic loader over the test split.
      :rtype: DataLoader

      :raises RuntimeError: If called before ``setup``.