:py:mod:`kosmos.ml.config.factories.lr_scheduler`
=================================================

.. py:module:: kosmos.ml.config.factories.lr_scheduler







Classes
-------

.. py:class:: LearningRateSchedulerConfig

   Bases: :py:class:`abc.ABC`

   Learning rate scheduler configuration.





   |

   .. rubric:: Methods

   .. py:method:: get_instance(optimizer: torch.optim.Optimizer) -> torch.optim.lr_scheduler.LRScheduler

      Get the learning rate scheduler instance.

      :returns: Learning rate scheduler instance.
      :rtype: LRScheduler




----

.. py:class:: StepLearningRateSchedulerConfig(step_size: int, gamma: float = 0.1)

   Bases: :py:class:`LearningRateSchedulerConfig`

   Step learning rate scheduler configuration.

   Initialize the step learning rate scheduler configuration.

   :param step_size: Period of learning rate decay.
   :type step_size: int
   :param gamma: Multiplicative factor of learning rate decay. Defaults to 0.1.
   :type gamma: float





   |

   .. rubric:: Methods

   .. py:method:: get_instance(optimizer: torch.optim.Optimizer) -> torch.optim.lr_scheduler.StepLR

      Get the step learning rate scheduler instance.

      :param optimizer: Optimizer instance.
      :type optimizer: Optimizer

      :returns: Step learning rate scheduler instance.
      :rtype: StepLR




----

.. py:class:: ExponentialLearningRateSchedulerConfig(gamma: float)

   Bases: :py:class:`LearningRateSchedulerConfig`

   Exponential learning rate scheduler configuration.

   Initialize the exponential learning rate scheduler configuration.

   :param gamma: Multiplicative factor of learning rate decay.
   :type gamma: float





   |

   .. rubric:: Methods

   .. py:method:: get_instance(optimizer: torch.optim.Optimizer) -> torch.optim.lr_scheduler.ExponentialLR

      Get the exponential learning rate scheduler instance.

      :param optimizer: Optimizer instance.
      :type optimizer: Optimizer

      :returns: Exponential learning rate scheduler instance.
      :rtype: ExponentialLR




----

.. py:class:: CosineLearningRateSchedulerConfig(max_epochs: int, min_lr: float = 0.0)

   Bases: :py:class:`LearningRateSchedulerConfig`

   Cosine annealing learning rate scheduler configuration.

   Initialize the cosine learning rate scheduler configuration.

   :param max_epochs: Maximum number of epochs (iterations for the scheduler).
   :type max_epochs: int
   :param min_lr: Minimum learning rate. Defaults to 0.0.
   :type min_lr: float





   |

   .. rubric:: Methods

   .. py:method:: get_instance(optimizer: torch.optim.Optimizer) -> torch.optim.lr_scheduler.CosineAnnealingLR

      Get the cosine learning rate scheduler instance.

      :param optimizer: Optimizer instance.
      :type optimizer: Optimizer

      :returns: Cosine learning rate scheduler instance.
      :rtype: CosineAnnealingLR