// META: title=test WebNN API resample2d operation
// META: global=window,dedicatedworker
// META: variant=?cpu
// META: variant=?gpu
// META: variant=?npu
// META: script=../resources/utils.js
// META: timeout=long
'use strict';
// https://www.w3.org/TR/webnn/#api-mlgraphbuilder-resample2d-method
// Resample the tensor values from the source to the destination spatial
// dimensions according to the scaling factors.
//
// enum MLInterpolationMode {
// "nearest-neighbor",
// "linear"
// };
//
// dictionary MLResample2dOptions {
// MLInterpolationMode mode = "nearest-neighbor";
// sequence<float> scales;
// sequence<[EnforceRange] unsigned long> sizes;
// sequence<[EnforceRange] unsigned long> axes;
// };
//
// MLOperand resample2d(
// MLOperand input, optional MLResample2dOptions options = {});
const getResample2dPrecisionTolerance = (graphResources) => {
const args = graphResources.operators[0].arguments;
const options =
args.length === 2 ? {...args[1][Object.keys(args[1])[0]]} : {};
const expectedOutputs = graphResources.expectedOutputs;
const dataType =
expectedOutputs[Object.keys(expectedOutputs)[0]].descriptor.dataType;
let tolerance;
if (options.mode && options.mode === 'linear') {
// interpolation mode is linear
if (dataType === 'float32') {
tolerance = 84;
} else if (dataType === 'float16') {
tolerance = 10;
} else {
tolerance = 1;
}
} else {
// interpolation mode is nearest-neighbor
tolerance = 0;
}
return {metricType: 'ULP', value: tolerance};
};
const resample2dTests = [
{
'name': 'resample2d float32 4D tensor default options',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
3.8600528240203857, 45.18463134765625, 87.67153930664062,
98.7821044921875, 66.3741455078125, 3.411583423614502,
86.14930725097656, 95.98133850097656, 76.87126159667969,
16.52591323852539, 65.98783111572266, 25.470922470092773,
22.56010627746582, 92.08479309082031, 85.80876922607422,
92.63166046142578, 29.916208267211914, 75.40460968017578,
62.06375503540039, 1.7712159156799316, 99.4723129272461,
11.440549850463867, 25.396343231201172, 67.0217514038086
],
'descriptor': {'dimensions': [1, 1, 4, 6], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments': [{'input': 'resample2dInput'}],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
3.8600528240203857, 45.18463134765625, 87.67153930664062,
98.7821044921875, 66.3741455078125, 3.411583423614502,
86.14930725097656, 95.98133850097656, 76.87126159667969,
16.52591323852539, 65.98783111572266, 25.470922470092773,
22.56010627746582, 92.08479309082031, 85.80876922607422,
92.63166046142578, 29.916208267211914, 75.40460968017578,
62.06375503540039, 1.7712159156799316, 99.4723129272461,
11.440549850463867, 25.396343231201172, 67.0217514038086
],
'descriptor': {'dimensions': [1, 1, 4, 6], 'dataType': 'float32'}
}
}
}
},
{
'name': 'resample2d(upsample) float32 4D tensor options.scales',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
59.92947006225586, 41.98918914794922, 66.39534759521484,
90.7006607055664, 86.95105743408203, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 2, 3], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments':
[{'input': 'resample2dInput'}, {'options': {'scales': [2, 2]}}],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 4, 6], 'dataType': 'float32'}
}
}
}
},
{
'name': 'resample2d(upsample) float32 4D tensor options.sizes',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
59.92947006225586, 41.98918914794922, 66.39534759521484,
90.7006607055664, 86.95105743408203, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 2, 3], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments':
[{'input': 'resample2dInput'}, {'options': {'sizes': [4, 6]}}],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 4, 6], 'dataType': 'float32'}
}
}
}
},
{
'name':
'resample2d(upsample) float32 4D tensor options.sizes ignored options.scales',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
59.92947006225586, 41.98918914794922, 66.39534759521484,
90.7006607055664, 86.95105743408203, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 2, 3], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments': [
{'input': 'resample2dInput'},
{'options': {'scales': [0.5, 0.5], 'sizes': [4, 6]}}
],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 4, 6], 'dataType': 'float32'}
}
}
}
},
{
'name': 'resample2d(upsample) float32 4D tensor options.axes=[1, 2]',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
59.92947006225586, 41.98918914794922, 66.39534759521484,
90.7006607055664, 86.95105743408203, 79.10005187988281
],
'descriptor': {'dimensions': [1, 2, 3, 1], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments': [
{'input': 'resample2dInput'},
{'options': {'sizes': [4, 6], 'axes': [1, 2]}}
],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281
],
'descriptor': {'dimensions': [1, 4, 6, 1], 'dataType': 'float32'}
}
}
}
},
{
'name':
'resample2d(upsample) float32 4D tensor explicit options.axes=[2, 3]',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
59.92947006225586, 41.98918914794922, 66.39534759521484,
90.7006607055664, 86.95105743408203, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 2, 3], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments': [
{'input': 'resample2dInput'},
{'options': {'sizes': [4, 6], 'axes': [2, 3]}}
],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 4, 6], 'dataType': 'float32'}
}
}
}
},
{
'name':
'resample2d(upsample) float32 4D tensor explicit options.axes=[3, 2]',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
59.92947006225586, 41.98918914794922, 66.39534759521484,
90.7006607055664, 86.95105743408203, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 2, 3], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments': [
{'input': 'resample2dInput'},
{'options': {'sizes': [6, 4], 'axes': [3, 2]}}
],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 4, 6], 'dataType': 'float32'}
}
}
}
},
{
'name':
'resample2d(upsample) float32 4D tensor explicit options.mode=\'nearest-neighbor\'',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
59.92947006225586, 41.98918914794922, 66.39534759521484,
90.7006607055664, 86.95105743408203, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 2, 3], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments': [
{'input': 'resample2dInput'},
{'options': {'mode': 'nearest-neighbor', 'sizes': [4, 6]}}
],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
59.92947006225586, 59.92947006225586, 41.98918914794922,
41.98918914794922, 66.39534759521484, 66.39534759521484,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281,
90.7006607055664, 90.7006607055664, 86.95105743408203,
86.95105743408203, 79.10005187988281, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 4, 6], 'dataType': 'float32'}
}
}
}
},
{
'name':
'resample2d(upsample) float32 4D tensor options.scales options.mode=\'linear\'',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
59.92947006225586, 41.98918914794922, 66.39534759521484,
90.7006607055664, 86.95105743408203, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 2, 3], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments': [
{'input': 'resample2dInput'},
{'options': {'mode': 'linear', 'scales': [2, 2]}}
],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
59.92947006225586, 55.444400787353516, 46.47425842285156,
48.090728759765625, 60.29380798339844, 66.39534759521484,
67.62226867675781, 64.02411651611328, 56.82780838012695,
57.31512451171875, 65.48605346679688, 69.57152557373047,
83.00786590576172, 81.18354797363281, 77.534912109375,
75.76390838623047, 75.87055206298828, 75.92387390136719,
90.7006607055664, 89.76325988769531, 87.88845825195312,
84.9883041381836, 81.06280517578125, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 4, 6], 'dataType': 'float32'}
}
}
}
},
{
'name':
'resample2d(upsample) float32 4D tensor options.sizes options.mode=\'linear\'',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
59.92947006225586, 41.98918914794922, 66.39534759521484,
90.7006607055664, 86.95105743408203, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 2, 3], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments': [
{'input': 'resample2dInput'},
{'options': {'mode': 'linear', 'sizes': [4, 6]}}
],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
59.92947006225586, 55.444400787353516, 46.47425842285156,
48.090728759765625, 60.29380798339844, 66.39534759521484,
67.62226867675781, 64.02411651611328, 56.82780838012695,
57.31512451171875, 65.48605346679688, 69.57152557373047,
83.00786590576172, 81.18354797363281, 77.534912109375,
75.76390838623047, 75.87055206298828, 75.92387390136719,
90.7006607055664, 89.76325988769531, 87.88845825195312,
84.9883041381836, 81.06280517578125, 79.10005187988281
],
'descriptor': {'dimensions': [1, 1, 4, 6], 'dataType': 'float32'}
}
}
}
},
{
'name':
'resample2d(upsample) float32 4D tensor options.axes=[1, 2] options.mode=\'linear\'',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
59.92947006225586, 41.98918914794922, 66.39534759521484,
90.7006607055664, 86.95105743408203, 79.10005187988281
],
'descriptor': {'dimensions': [1, 2, 3, 1], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments': [
{'input': 'resample2dInput'},
{'options': {'mode': 'linear', 'sizes': [4, 6], 'axes': [1, 2]}}
],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
59.92947006225586, 55.444400787353516, 46.47425842285156,
48.090728759765625, 60.29380798339844, 66.39534759521484,
67.62226867675781, 64.02411651611328, 56.82780838012695,
57.31512451171875, 65.48605346679688, 69.57152557373047,
83.00786590576172, 81.18354797363281, 77.534912109375,
75.76390838623047, 75.87055206298828, 75.92387390136719,
90.7006607055664, 89.76325988769531, 87.88845825195312,
84.9883041381836, 81.06280517578125, 79.10005187988281
],
'descriptor': {'dimensions': [1, 4, 6, 1], 'dataType': 'float32'}
}
}
}
},
{
'name': 'resample2d(upsample) float32 4D tensor options.axes=[0, 1]',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
59.92947006225586, 90.7006607055664, 41.98918914794922,
86.95105743408203, 66.39534759521484, 79.10005187988281
],
'descriptor': {'dimensions': [3, 2, 1, 1], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments': [
{'input': 'resample2dInput'},
{'options': {'sizes': [6, 4], 'axes': [0, 1]}}
],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
59.92947006225586, 59.92947006225586, 90.7006607055664,
90.7006607055664, 59.92947006225586, 59.92947006225586,
90.7006607055664, 90.7006607055664, 41.98918914794922,
41.98918914794922, 86.95105743408203, 86.95105743408203,
41.98918914794922, 41.98918914794922, 86.95105743408203,
86.95105743408203, 66.39534759521484, 66.39534759521484,
79.10005187988281, 79.10005187988281, 66.39534759521484,
66.39534759521484, 79.10005187988281, 79.10005187988281
],
'descriptor': {'dimensions': [6, 4, 1, 1], 'dataType': 'float32'}
}
}
}
},
{
'name': 'resample2d(upsample) float32 4D tensor options.axes=[1, 0]',
'graph': {
'inputs': {
'resample2dInput': {
'data': [
59.92947006225586, 90.7006607055664, 41.98918914794922,
86.95105743408203, 66.39534759521484, 79.10005187988281
],
'descriptor': {'dimensions': [3, 2, 1, 1], 'dataType': 'float32'}
}
},
'operators': [{
'name': 'resample2d',
'arguments': [
{'input': 'resample2dInput'},
{'options': {'sizes': [4, 6], 'axes': [1, 0]}}
],
'outputs': 'resample2dOutput'
}],
'expectedOutputs': {
'resample2dOutput': {
'data': [
59.92947006225586, 59.92947006225586, 90.7006607055664,
90.7006607055664, 59.92947006225586, 59.92947006225586,
90.7006607055664, 90.7006607055664, 41.98918914794922,
41.98918914794922, 86.95105743408203, 86.95105743408203,
41.98918914794922, 41.98918914794922, 86.95105743408203,
86.95105743408203, 66.39534759521484, 66.39534759521484,
79.10005187988281, 79.10005187988281, 66.39534759521484,
66.39534759521484, 79.10005187988281, 79.10005187988281
],
'descriptor': {'dimensions': [6, 4, 1, 1], 'dataType': 'float32'}
}
}
}
}
];
if (navigator.ml) {
resample2dTests.forEach((test) => {
webnn_conformance_test(
buildGraphAndCompute, getResample2dPrecisionTolerance, test);
});
} else {
test(() => assert_implements(navigator.ml, 'missing navigator.ml'));
}
Codebase Browser
chromium