:py:mod:`kosmos.ml.config.sl_train`
===================================

.. py:module:: kosmos.ml.config.sl_train







Classes
-------

.. py:class:: SLTrainConfig


   Supervised learning training configuration.

   .. attribute:: dataset

      Supervised learning dataset.

      :type: SLDataset

   .. attribute:: train_split

      Fraction of dataset for training. Test split is 1 - train_split.

      :type: float

   .. attribute:: batch_size

      Number of samples per batch.

      :type: int

   .. attribute:: num_epochs

      Number of training epochs.

      :type: int

   .. attribute:: model_config

      Model configuration.

      :type: ModelConfig

   .. attribute:: optimizer_config

      Optimizer configuration.

      :type: OptimizerConfig

   .. attribute:: lr_scheduler_config

      Learning rate scheduler
      configuration. Defaults to None.

      :type: LearningRateSchedulerConfig | None

   .. attribute:: max_grad_norm

      Maximum gradient norm. Defaults to 1.0.

      :type: float | None

   .. attribute:: loss_config

      Loss function configuration.

      :type: LossConfig







----

.. py:class:: FLTrainConfig

   Bases: :py:class:`SLTrainConfig`

   Federated learning training configuration.

   .. attribute:: dataset

      Supervised learning dataset.

      :type: SLDataset

   .. attribute:: train_split

      Fraction of dataset for training. Test split is 1 - train_split.

      :type: float

   .. attribute:: batch_size

      Number of samples per batch.

      :type: int

   .. attribute:: num_epochs

      Number of training epochs.

      :type: int

   .. attribute:: model_config

      Model configuration.

      :type: ModelConfig

   .. attribute:: optimizer_config

      Optimizer configuration.

      :type: OptimizerConfig

   .. attribute:: lr_scheduler_config

      Learning rate scheduler
      configuration. Defaults to None.

      :type: LearningRateSchedulerConfig | None

   .. attribute:: max_grad_norm

      Maximum gradient norm. Defaults to 1.0.

      :type: float | None

   .. attribute:: loss_config

      Loss function configuration.

      :type: LossConfig

   .. attribute:: num_rounds

      Number of federated learning rounds.

      :type: int