all_reduce¶

dragon.distributed.all_reduce(
  inputs,
  reduction='mean',
  group=None,
  **kwargs
)[source]¶

Reduce the input across all nodes in a group.

Parameters:

• inputs (dragon.Tensor) – The input tensor.
• reduction (str, optional) – The reduction method.
• group (ProcessGroup, optional) – The group for communication.

Returns:

dragon.Tensor – The output tensor.