HorovodPlugin

class pytorch_lightning.plugins.training_type.HorovodPlugin(parallel_devices=None)

Bases: pytorch_lightning.plugins.training_type.parallel.ParallelPlugin

Plugin for Horovod distributed training integration.

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

Perform a all_gather on all processes

Return type

Tensor

barrier(*args, **kwargs)

Forces all possibly joined processes to wait for each other

broadcast(obj, src=0)

Broadcasts an object to all processes

Return type

object

model_to_device()

Moves the model to the correct device

post_backward(closure_loss)

Run after precision plugin executes backward

Return type

None

pre_dispatch()

Hook to do something before the training/evaluation/prediction starts.

reduce(tensor, group=None, reduce_op='mean')

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.

Returns

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

setup(model)

Called by the accelerator to finish setup.

property root_device

Returns the root device