RoundsDataset

class bayesflow.datasets.RoundsDataset(simulator: Simulator, batch_size: int, num_batches: int, epochs_per_round: int, adapter: Adapter | None, *, stage: str = 'training', **kwargs)

Bases: PyDataset

A dataset that is generated on-the-fly at the beginning of every n-th epoch.

property num_batches: int

Number of batches in the PyDataset.

Returns:

The number of batches in the PyDataset or None to indicate that the dataset is infinite.

on_epoch_end() → None

Method called at the end of every epoch.

regenerate() → None

Sample new batches of data from the joint distribution unconditionally

property max_queue_size

on_epoch_begin()

Method called at the beginning of every epoch.

property use_multiprocessing

property workers