TPUBf16PrecisionPlugin

class pytorch_lightning.plugins.precision.TPUBf16PrecisionPlugin(*args, **kwargs)

Bases: pytorch_lightning.plugins.precision.tpu.TPUPrecisionPlugin

Plugin that enables bfloats on TPUs.

connect(model, optimizers, lr_schedulers)

Connects this plugin to the accelerator and the training process.

Return type

Tuple[Module, List[Optimizer], List[Any]]

teardown()

This method is called to teardown the training process.

It is the right place to release memory and free other resources.

Return type

None