megengine.functional.zeros_like#
- zeros_like(inp)[source]#
Returns a tensor filled with zeros with the same shape and data type as input tensor.
- Parameters:
inp (Tensor) – input tensor from which to derive the output tensor shape.
- Return type:
- Returns:
a tensor having the same shape as input tensor and filled with zeros.
Examples
>>> x = F.arange(6, dtype='int32').reshape(2, 3) >>> F.zeros_like(x) Tensor([[0 0 0] [0 0 0]], dtype=int32, device=xpux:0)