helios.core.rng
===============

.. py:module:: helios.core.rng


Classes
-------

.. autoapisummary::

   helios.core.rng.DefaultNumpyRNG


Functions
---------

.. autoapisummary::

   helios.core.rng.get_default_seed
   helios.core.rng.create_default_numpy_rng
   helios.core.rng.get_default_numpy_rng
   helios.core.rng.seed_rngs
   helios.core.rng.get_rng_state_dict
   helios.core.rng.load_rng_state_dict


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

.. py:function:: get_default_seed() -> int

   Return the default seed.


.. py:class:: DefaultNumpyRNG(seed: int | list[int] | tuple[int] | None = None)

   Default RNG from Numpy.

   This is intended to serve as a replacement for the legacy random API from Numpy. The
   class wraps the new Generator instance, which is set to use PCG64. Functionality
   similar to the modules for PyTorch is provided for easy serialization and restoring.

   :param seed: (optional) the initial seed to use.


   .. py:property:: generator
      :type: numpy.random.Generator

      Return the Numpy Generator instance.


   .. py:method:: state_dict() -> collections.abc.Mapping[str, Any]

      Create a dictionary containing the RNG state.

      :returns: The state of the RNG.



   .. py:method:: load_state_dict(state_dict: collections.abc.Mapping[str, Any]) -> None

      Restore the RNG from the given state dictionary.

      :param state_dict: the state dictionary.



.. py:function:: create_default_numpy_rng(seed: int | list[int] | tuple[int] | None = None)

   Initialize the default RNG with the given seed.

   :param seed: (optional) the seed to use.


.. py:function:: get_default_numpy_rng() -> DefaultNumpyRNG

   Return the default RNG.

   :returns: The random generator.

   :raises RuntimeError: if the default Numpy RNG hasn't been created.


.. py:function:: seed_rngs(seed: int | None = None, skip_torch: bool = False) -> None

   Seed the default RNGs with the given seed.

   If no seed is given, then the default seed from Helios will be used. The RNGs that
   will be seeded are: PyTorch (+ CUDA if available), stdlib random, and the default
   Numpy generator.
   The ``skip_torch`` flag is intended to be used when seeding worker processes for
   dataloaders. In those cases, the RNGs for PyTorch have already been seeded, so we
   shouldn't be re-seeding them.

   :param seed: optional value to seed the random generators with.
   :param skip_torch: if True, torch RNGs won't be seeded.


.. py:function:: get_rng_state_dict() -> dict[str, Any]

   Get the state dict for the default RNGs.

   Default RNGs are: PyTorch (+ CUDA if available) and Random.

   :returns: The state of all RNGs.


.. py:function:: load_rng_state_dict(state_dict: dict[str, Any]) -> None

   Restore the default RNGs from the given state dict.

   See :py:func:`.get_rng_state_dict` for the list of default RNGs.

   :param state_dict: the state of the RNGs