share tensors within the same backend device to save memory #88
Description
At present, tensors with same content are allocated separately for each layer. This is a waste of memory.
{{ message }}
At present, tensors with same content are allocated separately for each layer. This is a waste of memory.