Compress#

Compress - 11 #

Version

name: Compress (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

Selects slices from an input tensor along a given axis where condition evaluates to True for each axis index. In case axis is not provided, input is flattened before elements are selected. Compress behaves like numpy.compress: https://docs.scipy.org/doc/numpy/reference/generated/numpy.compress.html

Attributes

axis: (Optional) Axis along which to take slices. If not specified, input is flattened before elements being selected. Negative value means counting dimensions from the back. Accepted range is [-r, r-1] where r = rank(input).

Inputs

input (heterogeneous) - T: Tensor of rank r >= 1.
condition (heterogeneous) - T1: Rank 1 tensor of booleans to indicate which slices or data elements to be selected. Its length can be less than the input length along the axis or the flattened input size if axis is not specified. In such cases data slices or elements exceeding the condition length are discarded.

Outputs

output (heterogeneous) - T: Tensor of rank r if axis is specified. Otherwise output is a Tensor of rank 1.

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.
T1 in ( tensor(bool) ): Constrain to boolean tensors.

Examples

_compress_0

import numpy as np
import onnx

node = onnx.helper.make_node(
    "Compress",
    inputs=["input", "condition"],
    outputs=["output"],
    axis=0,
)
input = np.array([[1, 2], [3, 4], [5, 6]]).astype(np.float32)
condition = np.array([0, 1, 1])
output = np.compress(condition, input, axis=0)
# print(output)
# [[ 3.  4.]
# [ 5.  6.]]

expect(
    node,
    inputs=[input, condition.astype(bool)],
    outputs=[output],
    name="test_compress_0",
)

_compress_1

import numpy as np
import onnx

node = onnx.helper.make_node(
    "Compress",
    inputs=["input", "condition"],
    outputs=["output"],
    axis=1,
)
input = np.array([[1, 2], [3, 4], [5, 6]]).astype(np.float32)
condition = np.array([0, 1])
output = np.compress(condition, input, axis=1)
# print(output)
# [[ 2.]
# [ 4.]
# [ 6.]]

expect(
    node,
    inputs=[input, condition.astype(bool)],
    outputs=[output],
    name="test_compress_1",
)

_compress_default_axis

import numpy as np
import onnx

node = onnx.helper.make_node(
    "Compress",
    inputs=["input", "condition"],
    outputs=["output"],
)
input = np.array([[1, 2], [3, 4], [5, 6]]).astype(np.float32)
condition = np.array([0, 1, 0, 0, 1])
output = np.compress(condition, input)
# print(output)
# [ 2., 5.]

expect(
    node,
    inputs=[input, condition.astype(bool)],
    outputs=[output],
    name="test_compress_default_axis",
)

_compress_negative_axis

import numpy as np
import onnx

node = onnx.helper.make_node(
    "Compress",
    inputs=["input", "condition"],
    outputs=["output"],
    axis=-1,
)
input = np.array([[1, 2], [3, 4], [5, 6]]).astype(np.float32)
condition = np.array([0, 1])
output = np.compress(condition, input, axis=-1)
# print(output)
# [[ 2.]
# [ 4.]
# [ 6.]]
expect(
    node,
    inputs=[input, condition.astype(bool)],
    outputs=[output],
    name="test_compress_negative_axis",
)

Compress - 9 vs 11

Compress - 9 #