oneflow.broadcast_tensors(*tensors)List of Tensors

The interface is consistent with PyTorch. The documentation is referenced from:

Broadcasts the given tensors according to broadcasting-semantics.


*tensors – any number of tensors of the same type


More than one element of a broadcasted tensor may refer to a single memory location. As a result, in-place operations (especially ones that are vectorized) may result in incorrect behavior. If you need to write to the tensors, please clone them first.


>>> import oneflow as flow
>>> x = flow.arange(3).view(1, 3)
>>> y = flow.arange(2).view(2, 1)
>>> a, b = flow.broadcast_tensors(x, y)
>>> a.size()
oneflow.Size([2, 3])
>>> a
tensor([[0, 1, 2],
        [0, 1, 2]], dtype=oneflow.int64)