:py:mod:`kosmos.ml.fl.fl_client`
================================

.. py:module:: kosmos.ml.fl.fl_client







Classes
-------

.. py:class:: FLClient(trainer: kosmos.ml.sl_trainer.SLTrainer, node: kosmos.topology.node.Node)


   Federated learning client.

   Initialize a federated learning client.

   :param trainer: The trainer used for local training.
   :type trainer: SLTrainer
   :param node: The node representing this client in the topology.
   :type node: Node





   |

   .. rubric:: Methods

   .. py:method:: get_model_state() -> dict

      Get the current model state of the client.

      :returns: A state_dict containing the model parameters.
      :rtype: dict


   .. py:method:: set_model_state(state_dict: dict) -> None

      Set the client's model state.

      :param state_dict: The state_dict to set the client's model state to.
      :type state_dict: dict


   .. py:method:: train(num_epochs: int, dataloader: torch.utils.data.DataLoader, fl_round: int) -> collections.abc.Iterator[kosmos.ml.sl_result.SLTrainIterationResult]

      Train the client's model on the given data.

      :param num_epochs: Number of epochs to run.
      :type num_epochs: int
      :param dataloader: Dataloader providing the local training data.
      :type dataloader: DataLoader
      :param fl_round: The federated learning round index for this training run.
      :type fl_round: int

      :returns: An iterator yielding one training result per epoch.
      :rtype: Iterator[SLTrainIterationResult]