Source code
Revision control
Copy as Markdown
Other Tools
Test Info:
- This WPT test may be referenced by the following Test IDs:
- /webnn/conformance_tests/reduce_sum.https.any.html?cpu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/reduce_sum.https.any.html?gpu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/reduce_sum.https.any.html?npu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/reduce_sum.https.any.worker.html?cpu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/reduce_sum.https.any.worker.html?gpu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/reduce_sum.https.any.worker.html?npu - WPT Dashboard Interop Dashboard
// META: title=test WebNN API reduction operations
// META: global=window,dedicatedworker
// META: variant=?cpu
// META: variant=?gpu
// META: variant=?npu
// META: script=../resources/utils.js
// META: timeout=long
'use strict';
// Reduce the input tensor along all dimensions, or along the axes specified in
// the axes array parameter.
//
// dictionary MLReduceOptions {
// sequence<[EnforceRange] unsigned long> axes;
// boolean keepDimensions = false;
// };
//
// MLOperand reduceSum(MLOperand input, optional MLReduceOptions options = {});
const getReductionOperatorsPrecisionTolerance = (graphResources) => {
return {
metricType: 'ULP',
value: getReducedElementCount(graphResources),
};
};
const reduceSumTests = [
{
'name': 'reduceSum float32 0D constant tensor default options',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [69.6038589477539],
'descriptor': {shape: [], dataType: 'float32'},
'constant': true
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [{'input': 'reduceSumInput'}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': 69.6038589477539,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 0D constant tensor empty axes',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [69.6038589477539],
'descriptor': {shape: [], dataType: 'float32'},
'constant': true
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [{'input': 'reduceSumInput'}, {'options': {'axes': []}}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': 69.6038589477539,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 1D constant tensor all positive default options',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [24], dataType: 'float32'},
'constant': true
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [{'input': 'reduceSumInput'}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': 1313.87939453125,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 1D tensor all positive default options',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [24], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [{'input': 'reduceSumInput'}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': 1313.87939453125,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 1D tensor all negative default options',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
-51.77016830444336, -34.46467971801758, -40.98350524902344,
-83.34922790527344, -67.67525482177734, -18.7031192779541,
-20.28106117248535, -20.12305450439453, -83.63451385498047,
-23.651084899902344, -10.208438873291016, -36.2129020690918,
-76.26201629638672, -9.094745635986328, -53.889339447021484,
-67.52340698242188, -71.14580535888672, -82.04484558105469,
-96.29924774169922, -68.46700286865234, -26.107192993164062,
-68.0182113647461, -4.8330769538879395, -48.900699615478516
],
'descriptor': {shape: [24], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [{'input': 'reduceSumInput'}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': -1163.642578125,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 1D tensor all positive integers default options',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
56, 90, 67, 33, 20, 58, 22, 15, 86, 79, 59, 99,
16, 95, 67, 11, 60, 89, 50, 57, 77, 89, 10, 2
],
'descriptor': {shape: [24], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [{'input': 'reduceSumInput'}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput':
{'data': 1307, 'descriptor': {shape: [], dataType: 'float32'}}
}
}
},
{
'name': 'reduceSum float32 1D tensor all negative integers default options',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
-55, -36, -74, -17, -67, -95, -3, -67, -95, -13, -45, -9,
-33, -98, -86, -11, -70, -44, -31, -68, -79, -24, -60, -36
],
'descriptor': {shape: [24], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [{'input': 'reduceSumInput'}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput':
{'data': -1216, 'descriptor': {shape: [], dataType: 'float32'}}
}
}
},
{
'name': 'reduceSum float32 2D tensor default options',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [4, 6], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [{'input': 'reduceSumInput'}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': 1313.87939453125,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 3D tensor default options',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [2, 3, 4], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [{'input': 'reduceSumInput'}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': 1313.87939453125,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 4D tensor default options',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [{'input': 'reduceSumInput'}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': 1313.87939453125,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 5D tensor default options',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [2, 1, 4, 1, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [{'input': 'reduceSumInput'}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': 1313.87939453125,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 3D tensor options.axes',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [2, 3, 4], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [{'input': 'reduceSumInput'}, {'options': {'axes': [2]}}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': [
210.44241333007812, 104.54179382324219, 226.2177734375,
239.025390625, 260.405029296875, 273.2470397949219
],
'descriptor': {shape: [2, 3], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 4D tensor options.axes',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments':
[{'input': 'reduceSumInput'}, {'options': {'axes': [0, 2]}}],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': [
179.63900756835938, 260.37457275390625, 219.3611297607422,
246.83712768554688, 157.4895782470703, 250.1780242919922
],
'descriptor': {shape: [2, 3], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 3D tensor options.keepDimensions=false',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [2, 3, 4], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [
{'input': 'reduceSumInput'}, {'options': {'keepDimensions': false}}
],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': 1313.87939453125,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 3D tensor options.keepDimensions=true',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [2, 3, 4], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [
{'input': 'reduceSumInput'}, {'options': {'keepDimensions': true}}
],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': [1313.87939453125],
'descriptor': {shape: [1, 1, 1], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 4D tensor options.keepDimensions=false',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [
{'input': 'reduceSumInput'}, {'options': {'keepDimensions': false}}
],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': 1313.87939453125,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceSum float32 4D tensor options.keepDimensions=true',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [
{'input': 'reduceSumInput'}, {'options': {'keepDimensions': true}}
],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': [1313.87939453125],
'descriptor': {shape: [1, 1, 1, 1], dataType: 'float32'}
}
}
}
},
{
'name':
'reduceSum float32 4D tensor options.axes with options.keepDimensions=false',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [
{'input': 'reduceSumInput'},
{'options': {'axes': [1, 3], 'keepDimensions': false}}
],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': [
355.21942138671875, 185.98255920410156, 362.3598937988281,
410.3175354003906
],
'descriptor': {shape: [2, 2], dataType: 'float32'}
}
}
}
},
{
'name':
'reduceSum float32 4D tensor options.axes with options.keepDimensions=true',
'graph': {
'inputs': {
'reduceSumInput': {
'data': [
69.6038589477539, 99.17485809326172, 32.78234100341797,
8.881362915039062, 16.094295501708984, 11.80689525604248,
32.64223861694336, 43.99836349487305, 77.01776885986328,
79.79425811767578, 45.00794982910156, 24.397796630859375,
57.502685546875, 57.60173034667969, 80.26985931396484,
43.65110778808594, 87.5036849975586, 94.50203704833984,
35.54289627075195, 42.856414794921875, 88.58631896972656,
98.85772705078125, 25.626853942871094, 60.1761360168457
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceSum',
'arguments': [
{'input': 'reduceSumInput'},
{'options': {'axes': [1, 3], 'keepDimensions': true}}
],
'outputs': 'reduceSumOutput'
}],
'expectedOutputs': {
'reduceSumOutput': {
'data': [
355.21942138671875, 185.98255920410156, 362.3598937988281,
410.3175354003906
],
'descriptor': {shape: [2, 1, 2, 1], dataType: 'float32'}
}
}
}
}
];
if (navigator.ml) {
reduceSumTests.forEach((test) => {
webnn_conformance_test(
buildAndExecuteGraph, getReductionOperatorsPrecisionTolerance, test);
});
} else {
test(() => assert_implements(navigator.ml, 'missing navigator.ml'));
}