megengine.functional.reshape¶
- reshape(inp, target_shape)[源代码]¶
Reshapes a tensor without changing its data.
- 参数
inp (
Tensor) – input tensor to reshape.target_shape (
Iterable[int]) – target shape compatible with the original shape. One shape dimension is allowed to be -1 . When a shape dimension is -1 , the corresponding output tensor shape dimension must be inferred from the length of the tensor and the remaining dimensions.
- 返回类型
- 返回
an output tensor having the same data type, elements, and underlying element order as inp .
实际案例
>>> x = F.arange(12) >>> x Tensor([ 0. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.], device=xpux:0) >>> F.reshape(x, (3, 4)) Tensor([[ 0. 1. 2. 3.] [ 4. 5. 6. 7.] [ 8. 9. 10. 11.]], device=xpux:0) >>> F.reshape(x, (2, -1)) Tensor([[ 0. 1. 2. 3. 4. 5.] [ 6. 7. 8. 9. 10. 11.]], device=xpux:0)