# tf.contrib.nccl.all_sum

tf.contrib.nccl.all_sum(tensors)


Returns a list of tensors with the all-reduce sum across tensors.

The computation is done with an all-reduce operation, so if only some of the returned tensors are evaluated then the computation will hang.

#### Args:

• tensors: The input tensors across which to sum; must be assigned to GPU devices.

#### Returns:

List of tensors, each with the sum of the input tensors, where tensor i has the same device as tensors[i].