ReduceSum#

ReduceSum - 13 #

Version

name: ReduceSum (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

Computes the sum of the input tensor’s element along the provided axes. The resulting tensor has the same rank as the input if keepdims equals 1. If keepdims equals 0, then the resulting tensor has the reduced dimension pruned.

The above behavior is similar to numpy, with the exception that numpy defaults keepdims to False instead of True.

Attributes

keepdims: Keep the reduced dimension or not, default 1 means keep reduced dimension.
noop_with_empty_axes: Defines behaviour if ‘axes’ is empty. Default behaviour with ‘false’ is to reduce all axes. When axes is empty and this attribute is set to true, input tensor will not be reduced,and the output tensor would be equivalent to input tensor.

Inputs

Between 1 and 2 inputs.

data (heterogeneous) - T: An input tensor.
axes (optional, heterogeneous) - tensor(int64): Optional input list of integers, along which to reduce. The default is to reduce over all the dimensions of the input tensor if ‘noop_with_empty_axes’ is false, else act as an Identity op when ‘noop_with_empty_axes’ is true. Accepted range is [-r, r-1] where r = rank(data).

Outputs

reduced (heterogeneous) - T: Reduced output tensor.

Type Constraints

T in ( tensor(bfloat16), tensor(double), tensor(float), tensor(float16), tensor(int32), tensor(int64), tensor(uint32), tensor(uint64) ): Constrain input and output types to high-precision numeric tensors.

Examples

_do_not_keepdims

import numpy as np
import onnx

shape = [3, 2, 2]
axes = np.array([1], dtype=np.int64)
keepdims = 0

node = onnx.helper.make_node(
    "ReduceSum", inputs=["data", "axes"], outputs=["reduced"], keepdims=keepdims
)

data = np.array(
    [[[1, 2], [3, 4]], [[5, 6], [7, 8]], [[9, 10], [11, 12]]], dtype=np.float32
)
reduced = np.sum(data, axis=tuple(axes.tolist()), keepdims=keepdims == 1)
# print(reduced)
# [[4., 6.]
# [12., 14.]
# [20., 22.]]

expect(
    node,
    inputs=[data, axes],
    outputs=[reduced],
    name="test_reduce_sum_do_not_keepdims_example",
)

np.random.seed(0)
data = np.random.uniform(-10, 10, shape).astype(np.float32)
reduced = np.sum(data, axis=tuple(axes.tolist()), keepdims=keepdims == 1)

expect(
    node,
    inputs=[data, axes],
    outputs=[reduced],
    name="test_reduce_sum_do_not_keepdims_random",
)

_keepdims

import numpy as np
import onnx

shape = [3, 2, 2]
axes = np.array([1], dtype=np.int64)
keepdims = 1

node = onnx.helper.make_node(
    "ReduceSum", inputs=["data", "axes"], outputs=["reduced"], keepdims=keepdims
)

data = np.array(
    [[[1, 2], [3, 4]], [[5, 6], [7, 8]], [[9, 10], [11, 12]]], dtype=np.float32
)
reduced = np.sum(data, axis=tuple(axes.tolist()), keepdims=keepdims == 1)
# print(reduced)
# [[[4., 6.]]
# [[12., 14.]]
# [[20., 22.]]]

expect(
    node,
    inputs=[data, axes],
    outputs=[reduced],
    name="test_reduce_sum_keepdims_example",
)

np.random.seed(0)
data = np.random.uniform(-10, 10, shape).astype(np.float32)
reduced = np.sum(data, axis=tuple(axes.tolist()), keepdims=keepdims == 1)

expect(
    node,
    inputs=[data, axes],
    outputs=[reduced],
    name="test_reduce_sum_keepdims_random",
)

_default_axes_keepdims

import numpy as np
import onnx

shape = [3, 2, 2]
axes = np.array([], dtype=np.int64)
keepdims = 1

node = onnx.helper.make_node(
    "ReduceSum", inputs=["data", "axes"], outputs=["reduced"], keepdims=keepdims
)

data = np.array(
    [[[1, 2], [3, 4]], [[5, 6], [7, 8]], [[9, 10], [11, 12]]], dtype=np.float32
)
reduced = np.sum(data, axis=None, keepdims=keepdims == 1)
# print(reduced)
# [[[78.]]]

expect(
    node,
    inputs=[data, axes],
    outputs=[reduced],
    name="test_reduce_sum_default_axes_keepdims_example",
)

np.random.seed(0)
data = np.random.uniform(-10, 10, shape).astype(np.float32)
reduced = np.sum(data, axis=None, keepdims=keepdims == 1)

expect(
    node,
    inputs=[data, axes],
    outputs=[reduced],
    name="test_reduce_sum_default_axes_keepdims_random",
)

_negative_axes_keepdims

import numpy as np
import onnx

shape = [3, 2, 2]
axes = np.array([-2], dtype=np.int64)
keepdims = 1

node = onnx.helper.make_node(
    "ReduceSum", inputs=["data", "axes"], outputs=["reduced"], keepdims=keepdims
)

data = np.array(
    [[[1, 2], [3, 4]], [[5, 6], [7, 8]], [[9, 10], [11, 12]]], dtype=np.float32
)
reduced = np.sum(data, axis=tuple(axes.tolist()), keepdims=keepdims == 1)
# print(reduced)
# [[[4., 6.]]
# [[12., 14.]]
# [[20., 22.]]]

expect(
    node,
    inputs=[data, axes],
    outputs=[reduced],
    name="test_reduce_sum_negative_axes_keepdims_example",
)

np.random.seed(0)
data = np.random.uniform(-10, 10, shape).astype(np.float32)
reduced = np.sum(data, axis=tuple(axes.tolist()), keepdims=keepdims == 1)

expect(
    node,
    inputs=[data, axes],
    outputs=[reduced],
    name="test_reduce_sum_negative_axes_keepdims_random",
)

_empty_axes_input_noop

import numpy as np
import onnx

shape = [3, 2, 2]
keepdims = 1

node = onnx.helper.make_node(
    "ReduceSum",
    inputs=["data", "axes"],
    outputs=["reduced"],
    keepdims=keepdims,
    noop_with_empty_axes=True,
)

data = np.array(
    [[[1, 2], [3, 4]], [[5, 6], [7, 8]], [[9, 10], [11, 12]]], dtype=np.float32
)
axes = np.array([], dtype=np.int64)
reduced = np.array(data)
# print(reduced)
# [[[1, 2], [3, 4]], [[5, 6], [7, 8]], [[9, 10], [11, 12]]]

expect(
    node,
    inputs=[data, axes],
    outputs=[reduced],
    name="test_reduce_sum_empty_axes_input_noop_example",
)

np.random.seed(0)
data = np.random.uniform(-10, 10, shape).astype(np.float32)
reduced = np.array(data)

expect(
    node,
    inputs=[data, axes],
    outputs=[reduced],
    name="test_reduce_sum_negative_axes_keepdims_random",
)

ReduceSum - 11 vs 13

ReduceSum - 11 #

Version

name: ReduceSum (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

Computes the sum of the input tensor’s element along the provided axes. The resulting tensor has the same rank as the input if keepdims equals 1. If keepdims equal 0, then the resulted tensor have the reduced dimension pruned.

The above behavior is similar to numpy, with the exception that numpy defaults keepdims to False instead of True.

Attributes

axes: A list of integers, along which to reduce. The default is to reduce over all the dimensions of the input tensor. Accepted range is [-r, r-1] where r = rank(data).
keepdims: Keep the reduced dimension or not, default 1 means keep reduced dimension.

Inputs

data (heterogeneous) - T: An input tensor.

Outputs

reduced (heterogeneous) - T: Reduced output tensor.

Type Constraints

T in ( tensor(double), tensor(float), tensor(float16), tensor(int32), tensor(int64), tensor(uint32), tensor(uint64) ): Constrain input and output types to high-precision numeric tensors.

ReduceSum - 1 #

Version

name: ReduceSum (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

The above behavior is similar to numpy, with the exception that numpy defaults keepdims to False instead of True.

Attributes

axes: A list of integers, along which to reduce. The default is to reduce over all the dimensions of the input tensor.
keepdims: Keep the reduced dimension or not, default 1 means keep reduced dimension.

Inputs

data (heterogeneous) - T: An input tensor.

Outputs

reduced (heterogeneous) - T: Reduced output tensor.

Type Constraints

T in ( tensor(double), tensor(float), tensor(float16), tensor(int32), tensor(int64), tensor(uint32), tensor(uint64) ): Constrain input and output types to high-precision numeric tensors.

ReduceSum#

ReduceSum - 13#

ReduceSum - 11#

ReduceSum - 1#

ReduceSum - 13 #

ReduceSum - 11 #

ReduceSum - 1 #