Squeeze

Squeeze - 13

Version

• name: Squeeze (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

Remove single-dimensional entries from the shape of a tensor. Takes an input axes with a list of axes to squeeze. If axes is not provided, all the single dimensions will be removed from the shape. If an axis is selected with shape entry not equal to one, an error is raised.

Inputs

Between 1 and 2 inputs.

• data (heterogeneous) - T: Tensors with at least max(dims) dimensions.

• axes (optional, heterogeneous) - tensor(int64): List of integers indicating the dimensions to squeeze. Negative value means counting dimensions from the back. Accepted range is [-r, r-1] where r = rank(data).

Outputs

• squeezed (heterogeneous) - T: Reshaped tensor with same data as input.

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

squeeze

node = onnx.helper.make_node(
    'Squeeze',
    inputs=['x', 'axes'],
    outputs=['y'],
)
x = np.random.randn(1, 3, 4, 5).astype(np.float32)
axes = np.array([0], dtype=np.int64)
y = np.squeeze(x, axis=0)

expect(node, inputs=[x, axes], outputs=[y],
       name='test_squeeze')

squeeze_negative_axes

node = onnx.helper.make_node(
    'Squeeze',
    inputs=['x', 'axes'],
    outputs=['y'],
)
x = np.random.randn(1, 3, 1, 5).astype(np.float32)
axes = np.array([-2], dtype=np.int64)
y = np.squeeze(x, axis=-2)
expect(node, inputs=[x, axes], outputs=[y],
       name='test_squeeze_negative_axes')

Differences

0	0	Remove single-dimensional entries from the shape of a tensor.	Remove single-dimensional entries from the shape of a tensor.
1	1	Takes a parameter axes with a list of axes to squeeze.	Takes an input axes with a list of axes to squeeze.
2	2	If axes is not provided, all the single dimensions will be removed from	If axes is not provided, all the single dimensions will be removed from
3	3	the shape. If an axis is selected with shape entry not equal to one, an error is raised.	the shape. If an axis is selected with shape entry not equal to one, an error is raised.
4	4
	5		**Inputs**
	6
	7		Between 1 and 2 inputs.
	8
5	9	**Attributes**	* **data** (heterogeneous) - **T**:
6
	10		Tensors with at least max(dims) dimensions.
7	11	* **axes**:	* **axes** (optional, heterogeneous) - **tensor(int64)**:
8	12	List of integers indicating the dimensions to squeeze. Negative	List of integers indicating the dimensions to squeeze. Negative
9	13	value means counting dimensions from the back. Accepted range is	value means counting dimensions from the back. Accepted range is
10	14	[-r, r-1] where r = rank(data).	[-r, r-1] where r = rank(data).
11	15
12		**Inputs**
13
14		* **data** (heterogeneous) - **T**:
15		Tensors with at least max(dims) dimensions.
16
17	16	**Outputs**	**Outputs**
18	17
19	18	* **squeezed** (heterogeneous) - **T**:	* **squeezed** (heterogeneous) - **T**:
20	19	Reshaped tensor with same data as input.	Reshaped tensor with same data as input.
21	20
22	21	**Type Constraints**	**Type Constraints**
23	22
24	23	* **T** in (	* **T** in (
	24		tensor(bfloat16),
25	25	tensor(bool),	tensor(bool),
26	26	tensor(complex128),	tensor(complex128),
27	27	tensor(complex64),	tensor(complex64),
28	28	tensor(double),	tensor(double),
29	29	tensor(float),	tensor(float),
30	30	tensor(float16),	tensor(float16),
31	31	tensor(int16),	tensor(int16),
32	32	tensor(int32),	tensor(int32),
33	33	tensor(int64),	tensor(int64),
34	34	tensor(int8),	tensor(int8),
35	35	tensor(string),	tensor(string),
36	36	tensor(uint16),	tensor(uint16),
37	37	tensor(uint32),	tensor(uint32),
38	38	tensor(uint64),	tensor(uint64),
39	39	tensor(uint8)	tensor(uint8)
40	40	):	):
41	41	Constrain input and output types to all tensor types.	Constrain input and output types to all tensor types.

Squeeze - 11

Version

• name: Squeeze (GitHub)

• domain: main

• since_version: 11

• function: False

• support_level: SupportType.COMMON

• shape inference: True

This version of the operator has been available since version 11.

Summary

Remove single-dimensional entries from the shape of a tensor. Takes a parameter axes with a list of axes to squeeze. If axes is not provided, all the single dimensions will be removed from the shape. If an axis is selected with shape entry not equal to one, an error is raised.

Attributes

• axes: List of integers indicating the dimensions to squeeze. Negative value means counting dimensions from the back. Accepted range is [-r, r-1] where r = rank(data).

Inputs

• data (heterogeneous) - T: Tensors with at least max(dims) dimensions.

Outputs

• squeezed (heterogeneous) - T: Reshaped tensor with same data as input.

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.

Differences

0	0	Remove single-dimensional entries from the shape of a tensor.	Remove single-dimensional entries from the shape of a tensor.
1	1	Takes a parameter axes with a list of axes to squeeze.	Takes a parameter axes with a list of axes to squeeze.
2	2	If axes is not provided, all the single dimensions will be removed from	If axes is not provided, all the single dimensions will be removed from
3	3	the shape. If an axis is selected with shape entry not equal to one, an error is raised.	the shape. If an axis is selected with shape entry not equal to one, an error is raised.
4	4
5	5	**Attributes**	**Attributes**
6	6
7	7	* **axes**:	* **axes**:
8	8	List of non-negative integers, indicate the dimensions to squeeze.	List of integers indicating the dimensions to squeeze. Negative
	9		value means counting dimensions from the back. Accepted range is
	10		[-r, r-1] where r = rank(data).
9	11
10	12	**Inputs**	**Inputs**
11	13
12	14	* **data** (heterogeneous) - **T**:	* **data** (heterogeneous) - **T**:
13	15	Tensors with at least max(dims) dimensions.	Tensors with at least max(dims) dimensions.
14	16
15	17	**Outputs**	**Outputs**
16	18
17	19	* **squeezed** (heterogeneous) - **T**:	* **squeezed** (heterogeneous) - **T**:
18	20	Reshaped tensor with same data as input.	Reshaped tensor with same data as input.
19	21
20	22	**Type Constraints**	**Type Constraints**
21	23
22	24	* **T** in (	* **T** in (
23	25	tensor(bool),	tensor(bool),
24	26	tensor(complex128),	tensor(complex128),
25	27	tensor(complex64),	tensor(complex64),
26	28	tensor(double),	tensor(double),
27	29	tensor(float),	tensor(float),
28	30	tensor(float16),	tensor(float16),
29	31	tensor(int16),	tensor(int16),
30	32	tensor(int32),	tensor(int32),
31	33	tensor(int64),	tensor(int64),
32	34	tensor(int8),	tensor(int8),
33	35	tensor(string),	tensor(string),
34	36	tensor(uint16),	tensor(uint16),
35	37	tensor(uint32),	tensor(uint32),
36	38	tensor(uint64),	tensor(uint64),
37	39	tensor(uint8)	tensor(uint8)
38	40	):	):
39	41	Constrain input and output types to all tensor types.	Constrain input and output types to all tensor types.

Squeeze - 1

Version

• name: Squeeze (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

Remove single-dimensional entries from the shape of a tensor. Takes a parameter axes with a list of axes to squeeze. If axes is not provided, all the single dimensions will be removed from the shape. If an axis is selected with shape entry not equal to one, an error is raised.

Attributes

• axes: List of non-negative integers, indicate the dimensions to squeeze.

Inputs

• data (heterogeneous) - T: Tensors with at least max(dims) dimensions.

Outputs

• squeezed (heterogeneous) - T: Reshaped tensor with same data as input.

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

Sqrt

next

StringNormalizer