mindspore.ops.atleast_3d

mindspore.ops.atleast_3d(inputs)[source]

Reshapes Tensor in inputs, every Tensor has at least 3 dimension after this operation.

Scalar, 1-D or 2-D Tensor is converted to 3-D Tensor, tensor with higher dimensions will be returned as it is.

Parameters: inputs (Union[Tensor, list[Tensor]]) – One or more input tensors.
Returns: Tensor or list[Tensor]. If returned a list, every element a in that list satisfies a.ndim >= 3. For example, a 1-D Tensor of shape \((N,)\) becomes a Tensor of shape \((1, N, 1)\), and a 2-D Tensor of shape \((M, N)\) becomes a tensor of shape \((M, N, 1)\).
Raises: TypeError – If the inputs is not a tensor or a list of tensors.

Supported Platforms:: Ascend GPU CPU

Examples

>>> import numpy as np
>>> from mindspore import Tensor, ops
>>> x1 = Tensor(np.ones((2, 3)))
>>> x2 = Tensor(np.ones(()))
>>> x3 = Tensor(np.ones(5))
>>> out = ops.atleast_3d([x1, x2, x3])
>>> print(out[0].asnumpy())
[[[1.]
  [1.]
  [1.]]

 [[1.]
  [1.]
  [1.]]]
>>> print(out[1].asnumpy())
[[[1.]]]
>>> print(out[2].asnumpy())
[[[1.]
  [1.]
  [1.]
  [1.]
  [1.]]]