torch.foreach.max

torch.foreach.max(inputs: TensorList, /) → tuple[Tensor, ...]

Returns the maximum value of each tensor in inputs.

This is semantically equivalent to applying torch.max() independently at every list position. Does not mutate its arguments and returns a tuple of result tensors.

Tensor-list arguments must be non-empty. An accelerated multi-tensor implementation is used only when supported by the inputs; otherwise the operation falls back to per-tensor execution.

This operation reduces every input tensor over all dimensions. It does not accept a dimension or return indices.

Parameters:

inputs (list or tuple of Tensor) – tensors to reduce.

Returns:

a tuple of scalar tensors.