XLAStrategy

class lightning.pytorch.strategies.XLAStrategy(accelerator=None, parallel_devices=None, checkpoint_io=None, precision_plugin=None, debug=False, sync_module_states=True, **_)

Bases: DDPStrategy

Strategy for training multiple TPU devices using the torch_xla.distributed.xla_multiprocessing.spawn() method.

all_gather(tensor, group=None, sync_grads=False)

Function to gather a tensor from several distributed processes.

Parameters:

• tensor (Tensor) – tensor to all-gather.

• group (Optional[Any]) – unused.

• sync_grads (bool) – flag that allows users to synchronize gradients for the all-gather operation.

Return type:

Tensor

Returns:

A tensor of shape (world_size, …)

barrier(name=None, *args, **kwargs)

Synchronizes all processes which blocks processes until the whole group enters this function.

Parameters:

name (Optional[str]) – an optional name to pass into barrier.

Return type:

None

broadcast(obj, src=0)

Broadcasts an object to all processes.

Parameters:

• obj (TypeVar(TBroadcast)) – the object to broadcast

• src (int) – source rank

Return type:

TypeVar(TBroadcast)

model_to_device()

Moves the model to the correct device.

Return type:

None

process_dataloader(dataloader)

Wraps the dataloader if necessary.

Parameters:

dataloader (object) – iterable. Ideally of type: torch.utils.data.DataLoader

Return type:

MpDeviceLoader

reduce(output, group=None, reduce_op=None)

Reduces a tensor from several distributed processes to one aggregated tensor.

Parameters:

• tensor – the tensor to sync and reduce

• group (Optional[Any]) – the process group to gather results from. Defaults to all processes (world)

• reduce_op (Union[ReduceOp, str, None]) – the reduction operation. Defaults to ‘mean’/’avg’. Can also be a string ‘sum’ to calculate the sum during reduction.

Return type:

Tensor

Returns:

reduced value, except when the input was not a tensor the output remains is unchanged

remove_checkpoint(filepath)

Remove checkpoint filepath from the filesystem.

Parameters:

filepath (Union[str, Path]) – Path to checkpoint

Return type:

None

save_checkpoint(checkpoint, filepath, storage_options=None)

Save model/training states as a checkpoint file through state-dump and file-write.

Parameters:

• checkpoint (dict[str, Any]) – dict containing model and trainer state

• filepath (Union[str, Path]) – write-target file’s path

• storage_options (Optional[Any]) – parameter for how to save to storage, passed to CheckpointIO plugin

Return type:

None

setup(trainer)

Sets up the accelerator, plugins and initializes the optimizers (if needed).

Parameters:

trainer (Trainer) – the trainer instance

Return type:

None

setup_environment()

Setup any processes or distributed connections.

This is called before the LightningModule/DataModule setup hook which allows the user to access the accelerator environment before setup is complete.

Return type:

None

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

property root_device: device

Return the root device.