Transpose

Transpose - 13

Version

• name: Transpose (GitHub)

• domain: main

• since_version: 13

• function: False

• support_level: SupportType.COMMON

• shape inference: True

This version of the operator has been available since version 13.

Summary

Transpose the input tensor similar to numpy.transpose. For example, when perm=(1, 0, 2), given an input tensor of shape (1, 2, 3), the output shape will be (2, 1, 3).

Attributes

• perm: A list of integers. By default, reverse the dimensions, otherwise permute the axes according to the values given.

Inputs

• data (heterogeneous) - T: An input tensor.

Outputs

• transposed (heterogeneous) - T: Transposed output.

Type Constraints

• T in ( tensor(bfloat16), tensor(bool), tensor(complex128), tensor(complex64), tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(string), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8) ): Constrain input and output types to all tensor types.

Examples

default

shape = (2, 3, 4)
data = np.random.random_sample(shape).astype(np.float32)

node = onnx.helper.make_node(
    'Transpose',
    inputs=['data'],
    outputs=['transposed']
)

transposed = np.transpose(data)
expect(node, inputs=[data], outputs=[transposed],
       name='test_transpose_default')

all_permutations

shape = (2, 3, 4)
data = np.random.random_sample(shape).astype(np.float32)
permutations = list(itertools.permutations(np.arange(len(shape))))

for i in range(len(permutations)):
    node = onnx.helper.make_node(
        'Transpose',
        inputs=['data'],
        outputs=['transposed'],
        perm=permutations[i]
    )
    transposed = np.transpose(data, permutations[i])
    expect(node, inputs=[data], outputs=[transposed],
           name='test_transpose_all_permutations_' + str(i))

Differences

0	0	Transpose the input tensor similar to numpy.transpose. For example, when	Transpose the input tensor similar to numpy.transpose. For example, when
1	1	perm=(1, 0, 2), given an input tensor of shape (1, 2, 3), the output shape	perm=(1, 0, 2), given an input tensor of shape (1, 2, 3), the output shape
2	2	will be (2, 1, 3).	will be (2, 1, 3).
3	3
4	4	**Attributes**	**Attributes**
5	5
6	6	* **perm**:	* **perm**:
7	7	A list of integers. By default, reverse the dimensions, otherwise	A list of integers. By default, reverse the dimensions, otherwise
8	8	permute the axes according to the values given.	permute the axes according to the values given.
9	9
10	10	**Inputs**	**Inputs**
11	11
12	12	* **data** (heterogeneous) - **T**:	* **data** (heterogeneous) - **T**:
13	13	An input tensor.	An input tensor.
14	14
15	15	**Outputs**	**Outputs**
16	16
17	17	* **transposed** (heterogeneous) - **T**:	* **transposed** (heterogeneous) - **T**:
18	18	Transposed output.	Transposed output.
19	19
20	20	**Type Constraints**	**Type Constraints**
21	21
22	22	* **T** in (	* **T** in (
	23		tensor(bfloat16),
23	24	tensor(bool),	tensor(bool),
24	25	tensor(complex128),	tensor(complex128),
25	26	tensor(complex64),	tensor(complex64),
26	27	tensor(double),	tensor(double),
27	28	tensor(float),	tensor(float),
28	29	tensor(float16),	tensor(float16),
29	30	tensor(int16),	tensor(int16),
30	31	tensor(int32),	tensor(int32),
31	32	tensor(int64),	tensor(int64),
32	33	tensor(int8),	tensor(int8),
33	34	tensor(string),	tensor(string),
34	35	tensor(uint16),	tensor(uint16),
35	36	tensor(uint32),	tensor(uint32),
36	37	tensor(uint64),	tensor(uint64),
37	38	tensor(uint8)	tensor(uint8)
38	39	):	):
39	40	Constrain input and output types to all tensor types.	Constrain input and output types to all tensor types.

Transpose - 1

Version

• name: Transpose (GitHub)

• domain: main

• since_version: 1

• function: False

• support_level: SupportType.COMMON

• shape inference: True

This version of the operator has been available since version 1.

Summary

Transpose the input tensor similar to numpy.transpose. For example, when perm=(1, 0, 2), given an input tensor of shape (1, 2, 3), the output shape will be (2, 1, 3).

Attributes

• perm: A list of integers. By default, reverse the dimensions, otherwise permute the axes according to the values given.

Inputs

• data (heterogeneous) - T: An input tensor.

Outputs

• transposed (heterogeneous) - T: Transposed output.

Type Constraints

• T in ( tensor(bool), tensor(complex128), tensor(complex64), tensor(double), tensor(float), tensor(float16), tensor(int16), tensor(int32), tensor(int64), tensor(int8), tensor(string), tensor(uint16), tensor(uint32), tensor(uint64), tensor(uint8) ): Constrain input and output types to all tensor types.

previous

TopK

next

Trilu