lale.lib.lale.batching module

class lale.lib.lale.batching.BatchingImpl(operator=None, batch_size=32, shuffle=True, num_workers=0, inmemory=False)

Bases: object

Batching trains the given pipeline using batches. The batch_size is used across all steps of the pipeline, serializing the intermediate outputs if specified.

Parameters

• operator (lale.operators.Pipeline) – A Lale pipeline object that needs to be trained/used for transform or predictions, by default None

• batch_size (int, optional) – Batch size to be used for all steps in the pipeline, by default 32

• shuffle (bool, optional) – Shuffle dataset before batching or not, by default True

• num_workers (int, optional) – Number of workers for pytorch dataloader, by default 0

• inmemory (bool, optional) – Whether all the computations are done in memory or intermediate outputs are serialized.

fit(X, y=None)

predict(X, y=None)

transform(X, y=None)