gradient_accumulation_scheduler

Classes

GradientAccumulationScheduler

Change gradient accumulation factor according to scheduling.

Gradient Accumulator

Change gradient accumulation factor according to scheduling. Trainer also calls optimizer.step() for the last indivisible step number.

class pytorch_lightning.callbacks.gradient_accumulation_scheduler.GradientAccumulationScheduler(scheduling)

Bases: pytorch_lightning.callbacks.base.Callback

Change gradient accumulation factor according to scheduling.

Parameters

scheduling (Dict[int, int]) – scheduling in format {epoch: accumulation_factor}

Example:

>>> from pytorch_lightning import Trainer
>>> from pytorch_lightning.callbacks import GradientAccumulationScheduler

# at epoch 5 start accumulating every 2 batches
>>> accumulator = GradientAccumulationScheduler(scheduling={5: 2})
>>> trainer = Trainer(callbacks=[accumulator])

# alternatively, pass the scheduling dict directly to the Trainer
>>> trainer = Trainer(accumulate_grad_batches={5: 2})

on_epoch_start(trainer, pl_module)

Called when the epoch begins.