mindspore.ops.dsplit

mindspore.ops.dsplit(input, indices_or_sections)

Splits a tensor along the 3rd axis. It is equivalent to ops.tensor_split with $axis=2$ .

Parameters

• input (Tensor) – A tensor to be divided.

• indices_or_sections (Union[int, tuple(int), list(int)]) – See indices_or_sections argument in mindspore.ops.tensor_split().

Returns

Tuple of tensors.

Supported Platforms:

Ascend GPU CPU

Examples

>>> import mindspore
>>> input = mindspore.ops.arange(16.0).reshape(2, 2, 4)
>>> print(input)
[[[ 0.  1.  2.  3.]
  [ 4.  5.  6.  7.]]
 [[ 8.  9. 10. 11.]
  [12. 13. 14. 15.]]]
>>> output = mindspore.ops.dsplit(input, 2)
>>> print(output)
(Tensor(shape=[2, 2, 2], dtype=Float32, value=
[[[ 0.00000000e+00,  1.00000000e+00],
  [ 4.00000000e+00,  5.00000000e+00]],
 [[ 8.00000000e+00,  9.00000000e+00],
  [ 1.20000000e+01,  1.30000000e+01]]]), Tensor(shape=[2, 2, 2], dtype=Float32, value=
[[[ 2.00000000e+00,  3.00000000e+00],
  [ 6.00000000e+00,  7.00000000e+00]],
 [[ 1.00000000e+01,  1.10000000e+01],
  [ 1.40000000e+01,  1.50000000e+01]]]))
>>> output = mindspore.ops.dsplit(input, [3, 6])
>>> print(output)
(Tensor(shape=[2, 2, 3], dtype=Float32, value=
[[[ 0.00000000e+00,  1.00000000e+00,  2.00000000e+00],
  [ 4.00000000e+00,  5.00000000e+00,  6.00000000e+00]],
 [[ 8.00000000e+00,  9.00000000e+00,  1.00000000e+01],
  [ 1.20000000e+01,  1.30000000e+01,  1.40000000e+01]]]), Tensor(shape=[2, 2, 1], dtype=Float32, value=
[[[ 3.00000000e+00],
  [ 7.00000000e+00]],
 [[ 1.10000000e+01],
  [ 1.50000000e+01]]]), Tensor(shape=[2, 2, 0], dtype=Float32, value=
))