fl4health.clients.tabular_data_client module

class TabularDataClient(data_path, metrics, device, id_column, targets, loss_meter_type=LossMeterType.AVERAGE, checkpoint_and_state_module=None, reporters=None, progress_bar=False, client_name=None)[source]

Bases: BasicClient

__init__(data_path, metrics, device, id_column, targets, loss_meter_type=LossMeterType.AVERAGE, checkpoint_and_state_module=None, reporters=None, progress_bar=False, client_name=None)[source]

Client to facilitate federated feature space alignment, specifically for tabular data, and then perform federated training.

Parameters:

  • data_path (Path) – path to the data to be used to load the data for client-side training.

  • metrics (Sequence[Metric]) – Metrics to be computed based on the labels and predictions of the client model.

  • device (torch.device) – Device indicator for where to send the model, batches, labels etc. Often “cpu” or “cuda”.

  • id_column (str) – ID column. This is required for tabular encoding. It should be unique per row, but need not necessarily be a meaningful identifier (i.e. could be row number)

  • targets (str | list[str]) – The target column or columns name. This allows for multiple targets to be specified if desired.

  • loss_meter_type (LossMeterType, optional) – Type of meter used to track and compute the losses over each batch. Defaults to LossMeterType.AVERAGE.

  • checkpoint_and_state_module (ClientCheckpointAndStateModule | None, optional) – A module meant to handle both checkpointing and state saving. The module, and its underlying model and state checkpointing components will determine when and how to do checkpointing during client-side training. No checkpointing (state or model) is done if not provided. Defaults to None.

  • reporters (Sequence[BaseReporter] | None, optional) – A sequence of FL4Health reporters which the client should send data to. Defaults to None.

  • progress_bar (bool, optional) – Whether or not to display a progress bar during client training and validation. Uses tqdm. Defaults to False

  • client_name (str | None, optional) – An optional client name that uniquely identifies a client. If not passed, a hash is randomly generated. Client state will use this as part of its state file name. Defaults to None.

get_data_frame(config)[source]

User defined method that returns a pandas dataframe.

Parameters:

config (Config) – Configuration sent by the server for customization of the function

Return type:

DataFrame

get_properties(config)[source]

Return properties of client to be sent to the server. Depending on whether the server has communicated the information to be used for feature alignment, the client will send the input/output dimensions so the server can use them to initialize the global model.

First initializes the client because this is called prior to the first federated learning round.

Parameters:

config (Config) – Configuration sent by the server for customization of the function

Returns:

Properties to be returned to the server, providing information about the client.

Return type:

dict[str, Scalar]

preset_specific_pipeline(feature_name, pipeline)[source]

The user may use this method to specify a specific pipeline to be applied to a particular feature. This function stores the provided pipeline associated with the provided feature_name.

Parameters:

  • feature_name (str) – Name of the feature as a column in the dataframe

  • pipeline (Pipeline) – Pipeline of transformations to be applied to the target feature

Return type:

None

set_feature_specific_pipelines()[source]

Given the feature specific pipelines, at them to the tabular feature preprocessor.

Return type:

None

setup_client(config)[source]

Initialize the client by encoding the information of its tabular data and initializing the corresponding TabularFeaturesPreprocessor.

config[SOURCE_SPECIFIED] indicates whether the server has obtained the source of information to perform feature alignment. If it is True, it means the server has obtained such information (either a priori or by polling a client). So the client will encode that information and use it instead to perform feature preprocessing.

Parameters:

config (Config) – Configuration sent by the server for customization of the function

Return type:

None