Aggregator

Utility class for aggregating multiple dataloaders into a single iterable.

class _Aggregator(dataloaders, batching_mode)

Bases: object

Aggregate multiple dataloaders into a unified iterable object.

The aggregator combines batches produced by multiple dataloaders according to the selected batching strategy. It is primarily used to coordinate the iteration of multiple training conditions within a single training loop.

Initialization of the _Aggregator class.

Parameters:

• dataloaders (dict[str, DataLoader]) – The mapping between condition names and their corresponding dataloaders.

• batching_mode (str) – The strategy used to aggregate batches across dataloaders. Available options are "common_batch_size" for uniform batch sizes across conditions, "proportional" for batch sizes proportional to dataset sizes, and "separate_conditions" for iterating through each condition separately.

Raises:

NotImplementedError – If the selected batching mode is not yet implemented.