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_max.https.any.html?cpu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/reduce_max.https.any.html?gpu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/reduce_max.https.any.html?npu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/reduce_max.https.any.worker.html?cpu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/reduce_max.https.any.worker.html?gpu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/reduce_max.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 reduceMax(MLOperand input, optional MLReduceOptions options = {});
const getReductionOperatorsPrecisionTolerance = (graphResources) => {
return {
metricType: 'ULP',
value: 0,
};
};
const reduceMaxTests = [
{
'name': 'reduceMax float32 0D constant tensor default options',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [32.16658401489258],
'descriptor': {shape: [], dataType: 'float32'},
'constant': true
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [{'input': 'reduceMaxInput'}],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': 32.16658401489258,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 0D constant tensor empty axes',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [32.16658401489258],
'descriptor': {shape: [], dataType: 'float32'},
'constant': true
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [{'input': 'reduceMaxInput'}, {'options': {'axes': []}}],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': 32.16658401489258,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 1D constant tensor default options',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [24], dataType: 'float32'},
'constant': true
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [{'input': 'reduceMaxInput'}],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': 99.77313232421875,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 1D tensor default options',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [24], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [{'input': 'reduceMaxInput'}],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': 99.77313232421875,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 2D tensor default options',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [4, 6], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [{'input': 'reduceMaxInput'}],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': 99.77313232421875,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 3D tensor default options',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [2, 3, 4], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [{'input': 'reduceMaxInput'}],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': 99.77313232421875,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 4D tensor default options',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [{'input': 'reduceMaxInput'}],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': 99.77313232421875,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 5D tensor default options',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [2, 1, 4, 1, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [{'input': 'reduceMaxInput'}],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': 99.77313232421875,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 3D tensor options.axes',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [2, 3, 4], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [{'input': 'reduceMaxInput'}, {'options': {'axes': [2]}}],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': [
90.42288208007812, 75.90379333496094, 94.99645233154297,
96.55397033691406, 99.77313232421875, 20.253753662109375
],
'descriptor': {shape: [2, 3], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 4D tensor options.axes',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments':
[{'input': 'reduceMaxInput'}, {'options': {'axes': [0, 2]}}],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': [
62.504642486572266, 96.55397033691406, 99.77313232421875,
-21.557384490966797, 94.99645233154297, 37.28493118286133
],
'descriptor': {shape: [2, 3], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 3D tensor options.keepDimensions=false',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [2, 3, 4], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [
{'input': 'reduceMaxInput'}, {'options': {'keepDimensions': false}}
],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': 99.77313232421875,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 3D tensor options.keepDimensions=true',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [2, 3, 4], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [
{'input': 'reduceMaxInput'}, {'options': {'keepDimensions': true}}
],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': [99.77313232421875],
'descriptor': {shape: [1, 1, 1], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 4D tensor options.keepDimensions=false',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [
{'input': 'reduceMaxInput'}, {'options': {'keepDimensions': false}}
],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': 99.77313232421875,
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'reduceMax float32 4D tensor options.keepDimensions=true',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [
{'input': 'reduceMaxInput'}, {'options': {'keepDimensions': true}}
],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': [99.77313232421875],
'descriptor': {shape: [1, 1, 1, 1], dataType: 'float32'}
}
}
}
},
{
'name':
'reduceMax float32 4D tensor options.axes with options.keepDimensions=false',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [
{'input': 'reduceMaxInput'},
{'options': {'axes': [1, 3], 'keepDimensions': false}}
],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': [
90.42288208007812, 94.99645233154297, 96.55397033691406,
99.77313232421875
],
'descriptor': {shape: [2, 2], dataType: 'float32'}
}
}
}
},
{
'name':
'reduceMax float32 4D tensor options.axes with options.keepDimensions=true',
'graph': {
'inputs': {
'reduceMaxInput': {
'data': [
32.16658401489258, 90.42288208007812, -26.341794967651367,
-7.147959232330322, 75.90379333496094, -48.2042121887207,
-53.09425354003906, 66.66099548339844, -96.16854095458984,
-88.30545043945312, 94.99645233154297, 37.28493118286133,
-42.209861755371094, 96.55397033691406, 0.8807229995727539,
62.504642486572266, 36.650634765625, 99.77313232421875,
-72.86485290527344, -46.03200912475586, 20.253753662109375,
-21.557384490966797, -51.28727340698242, -42.58832931518555
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'reduceMax',
'arguments': [
{'input': 'reduceMaxInput'},
{'options': {'axes': [1, 3], 'keepDimensions': true}}
],
'outputs': 'reduceMaxOutput'
}],
'expectedOutputs': {
'reduceMaxOutput': {
'data': [
90.42288208007812, 94.99645233154297, 96.55397033691406,
99.77313232421875
],
'descriptor': {shape: [2, 1, 2, 1], dataType: 'float32'}
}
}
}
},
];
if (navigator.ml) {
reduceMaxTests.forEach((test) => {
webnn_conformance_test(
buildAndExecuteGraph, getReductionOperatorsPrecisionTolerance, test);
});
} else {
test(() => assert_implements(navigator.ml, 'missing navigator.ml'));
}