MixedPrecision

class lightning_fabric.plugins.precision.MixedPrecision(precision, device, scaler=None)

Bases: lightning_fabric.plugins.precision.precision.Precision

Plugin for Automatic Mixed Precision (AMP) training with torch.autocast.

Parameters:

• precision (Literal['16', 16, 'bf16']) – Whether to use torch.float16 (16) or torch.bfloat16 ('bf16').

• device (str) – The device for torch.autocast.

• scaler (Optional[GradScaler]) – An optional torch.cuda.amp.GradScaler to use.

backward(tensor, model, *args, **kwargs)

Performs the actual backpropagation.

Parameters:

• tensor (Tensor) – The tensor that will be used for backpropagation

• model (Optional[Module]) – The module that was involved in producing the tensor and whose parameters need the gradients

Return type:

None

convert_input(data)

Convert model inputs (forward) to the floating point precision type of this plugin.

This is a no-op for tensors that are not of floating-point type or already have the desired type.

Return type:

Tensor

forward_context()

A contextmanager for managing model forward/training_step/evaluation_step/predict_step.

Return type:

Generator[None, None, None]

load_state_dict(state_dict)

Called when loading a checkpoint, implement to reload precision plugin state given precision plugin state_dict.

Parameters:

state_dict (Dict[str, Any]) – the precision plugin state returned by state_dict.

Return type:

None

optimizer_step(optimizer, **kwargs)

Hook to run the optimizer step.

Return type:

Any

state_dict()

Called when saving a checkpoint, implement to generate precision plugin state_dict.

Return type:

Dict[str, Any]

Returns:

A dictionary containing precision plugin state.