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/sqrt.https.any.html?cpu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/sqrt.https.any.html?gpu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/sqrt.https.any.html?npu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/sqrt.https.any.worker.html?cpu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/sqrt.https.any.worker.html?gpu - WPT Dashboard Interop Dashboard
- /webnn/conformance_tests/sqrt.https.any.worker.html?npu - WPT Dashboard Interop Dashboard
// META: title=test WebNN API element-wise sqrt operation
// META: global=window,dedicatedworker
// META: variant=?cpu
// META: variant=?gpu
// META: variant=?npu
// META: script=../resources/utils.js
// META: timeout=long
'use strict';
// Compute the square root of the input tensor, element-wise.
//
// MLOperand sqrt(MLOperand input);
const getSqrtPrecisionTolerance = (graphResources) => {
const toleranceValueDict = {float32: 1, float16: 1};
const expectedDataType =
getExpectedDataTypeOfSingleOutput(graphResources.expectedOutputs);
return {metricType: 'ULP', value: toleranceValueDict[expectedDataType]};
};
const sqrtTests = [
{
'name': 'sqrt float32 0D scalar',
'graph': {
'inputs': {
'sqrtInput': {
'data': [4.0044636726379395],
'descriptor': {shape: [], dataType: 'float32'}
}
},
'operators': [{
'name': 'sqrt',
'arguments': [{'input': 'sqrtInput'}],
'outputs': 'sqrtOutput'
}],
'expectedOutputs': {
'sqrtOutput': {
'data': [2.001115560531616],
'descriptor': {shape: [], dataType: 'float32'}
}
}
}
},
{
'name': 'sqrt float32 1D constant tensor',
'graph': {
'inputs': {
'sqrtInput': {
'data': [
7.256007194519043, 7.786442279815674, 1.3684587478637695,
8.05341625213623, 9.131288528442383, 8.52578067779541,
4.870553493499756, 7.625959396362305, 2.705026865005493,
8.709602355957031, 3.2687935829162598, 4.712882995605469,
8.669181823730469, 8.829607009887695, 0.5529024600982666,
7.958771228790283, 4.09640645980835, 7.919884204864502,
4.424484729766846, 0.09894099831581116, 4.6900248527526855,
1.5277378559112549, 5.929779529571533, 6.066471576690674
],
'descriptor': {shape: [24], dataType: 'float32'},
'constant': true
}
},
'operators': [{
'name': 'sqrt',
'arguments': [{'input': 'sqrtInput'}],
'outputs': 'sqrtOutput'
}],
'expectedOutputs': {
'sqrtOutput': {
'data': [
2.693697690963745, 2.790419816970825, 1.1698113679885864,
2.8378541469573975, 3.0218021869659424, 2.919893980026245,
2.20693302154541, 2.7615139484405518, 1.644696593284607,
2.9512035846710205, 1.8079805374145508, 2.170917510986328,
2.944347381591797, 2.9714653491973877, 0.7435740828514099,
2.821129322052002, 2.023958206176758, 2.8142287731170654,
2.1034460067749023, 0.31454887986183167, 2.165646553039551,
1.2360169887542725, 2.4351139068603516, 2.4630208015441895
],
'descriptor': {shape: [24], dataType: 'float32'}
}
}
}
},
{
'name': 'sqrt float32 1D tensor',
'graph': {
'inputs': {
'sqrtInput': {
'data': [
7.256007194519043, 7.786442279815674, 1.3684587478637695,
8.05341625213623, 9.131288528442383, 8.52578067779541,
4.870553493499756, 7.625959396362305, 2.705026865005493,
8.709602355957031, 3.2687935829162598, 4.712882995605469,
8.669181823730469, 8.829607009887695, 0.5529024600982666,
7.958771228790283, 4.09640645980835, 7.919884204864502,
4.424484729766846, 0.09894099831581116, 4.6900248527526855,
1.5277378559112549, 5.929779529571533, 6.066471576690674
],
'descriptor': {shape: [24], dataType: 'float32'}
}
},
'operators': [{
'name': 'sqrt',
'arguments': [{'input': 'sqrtInput'}],
'outputs': 'sqrtOutput'
}],
'expectedOutputs': {
'sqrtOutput': {
'data': [
2.693697690963745, 2.790419816970825, 1.1698113679885864,
2.8378541469573975, 3.0218021869659424, 2.919893980026245,
2.20693302154541, 2.7615139484405518, 1.644696593284607,
2.9512035846710205, 1.8079805374145508, 2.170917510986328,
2.944347381591797, 2.9714653491973877, 0.7435740828514099,
2.821129322052002, 2.023958206176758, 2.8142287731170654,
2.1034460067749023, 0.31454887986183167, 2.165646553039551,
1.2360169887542725, 2.4351139068603516, 2.4630208015441895
],
'descriptor': {shape: [24], dataType: 'float32'}
}
}
}
},
{
'name': 'sqrt float32 2D tensor',
'graph': {
'inputs': {
'sqrtInput': {
'data': [
7.256007194519043, 7.786442279815674, 1.3684587478637695,
8.05341625213623, 9.131288528442383, 8.52578067779541,
4.870553493499756, 7.625959396362305, 2.705026865005493,
8.709602355957031, 3.2687935829162598, 4.712882995605469,
8.669181823730469, 8.829607009887695, 0.5529024600982666,
7.958771228790283, 4.09640645980835, 7.919884204864502,
4.424484729766846, 0.09894099831581116, 4.6900248527526855,
1.5277378559112549, 5.929779529571533, 6.066471576690674
],
'descriptor': {shape: [4, 6], dataType: 'float32'}
}
},
'operators': [{
'name': 'sqrt',
'arguments': [{'input': 'sqrtInput'}],
'outputs': 'sqrtOutput'
}],
'expectedOutputs': {
'sqrtOutput': {
'data': [
2.693697690963745, 2.790419816970825, 1.1698113679885864,
2.8378541469573975, 3.0218021869659424, 2.919893980026245,
2.20693302154541, 2.7615139484405518, 1.644696593284607,
2.9512035846710205, 1.8079805374145508, 2.170917510986328,
2.944347381591797, 2.9714653491973877, 0.7435740828514099,
2.821129322052002, 2.023958206176758, 2.8142287731170654,
2.1034460067749023, 0.31454887986183167, 2.165646553039551,
1.2360169887542725, 2.4351139068603516, 2.4630208015441895
],
'descriptor': {shape: [4, 6], dataType: 'float32'}
}
}
}
},
{
'name': 'sqrt float32 3D tensor',
'graph': {
'inputs': {
'sqrtInput': {
'data': [
7.256007194519043, 7.786442279815674, 1.3684587478637695,
8.05341625213623, 9.131288528442383, 8.52578067779541,
4.870553493499756, 7.625959396362305, 2.705026865005493,
8.709602355957031, 3.2687935829162598, 4.712882995605469,
8.669181823730469, 8.829607009887695, 0.5529024600982666,
7.958771228790283, 4.09640645980835, 7.919884204864502,
4.424484729766846, 0.09894099831581116, 4.6900248527526855,
1.5277378559112549, 5.929779529571533, 6.066471576690674
],
'descriptor': {shape: [2, 3, 4], dataType: 'float32'}
}
},
'operators': [{
'name': 'sqrt',
'arguments': [{'input': 'sqrtInput'}],
'outputs': 'sqrtOutput'
}],
'expectedOutputs': {
'sqrtOutput': {
'data': [
2.693697690963745, 2.790419816970825, 1.1698113679885864,
2.8378541469573975, 3.0218021869659424, 2.919893980026245,
2.20693302154541, 2.7615139484405518, 1.644696593284607,
2.9512035846710205, 1.8079805374145508, 2.170917510986328,
2.944347381591797, 2.9714653491973877, 0.7435740828514099,
2.821129322052002, 2.023958206176758, 2.8142287731170654,
2.1034460067749023, 0.31454887986183167, 2.165646553039551,
1.2360169887542725, 2.4351139068603516, 2.4630208015441895
],
'descriptor': {shape: [2, 3, 4], dataType: 'float32'}
}
}
}
},
{
'name': 'sqrt float32 4D tensor',
'graph': {
'inputs': {
'sqrtInput': {
'data': [
7.256007194519043, 7.786442279815674, 1.3684587478637695,
8.05341625213623, 9.131288528442383, 8.52578067779541,
4.870553493499756, 7.625959396362305, 2.705026865005493,
8.709602355957031, 3.2687935829162598, 4.712882995605469,
8.669181823730469, 8.829607009887695, 0.5529024600982666,
7.958771228790283, 4.09640645980835, 7.919884204864502,
4.424484729766846, 0.09894099831581116, 4.6900248527526855,
1.5277378559112549, 5.929779529571533, 6.066471576690674
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'sqrt',
'arguments': [{'input': 'sqrtInput'}],
'outputs': 'sqrtOutput'
}],
'expectedOutputs': {
'sqrtOutput': {
'data': [
2.693697690963745, 2.790419816970825, 1.1698113679885864,
2.8378541469573975, 3.0218021869659424, 2.919893980026245,
2.20693302154541, 2.7615139484405518, 1.644696593284607,
2.9512035846710205, 1.8079805374145508, 2.170917510986328,
2.944347381591797, 2.9714653491973877, 0.7435740828514099,
2.821129322052002, 2.023958206176758, 2.8142287731170654,
2.1034460067749023, 0.31454887986183167, 2.165646553039551,
1.2360169887542725, 2.4351139068603516, 2.4630208015441895
],
'descriptor': {shape: [2, 2, 2, 3], dataType: 'float32'}
}
}
}
},
{
'name': 'sqrt float32 5D tensor',
'graph': {
'inputs': {
'sqrtInput': {
'data': [
7.256007194519043, 7.786442279815674, 1.3684587478637695,
8.05341625213623, 9.131288528442383, 8.52578067779541,
4.870553493499756, 7.625959396362305, 2.705026865005493,
8.709602355957031, 3.2687935829162598, 4.712882995605469,
8.669181823730469, 8.829607009887695, 0.5529024600982666,
7.958771228790283, 4.09640645980835, 7.919884204864502,
4.424484729766846, 0.09894099831581116, 4.6900248527526855,
1.5277378559112549, 5.929779529571533, 6.066471576690674
],
'descriptor': {shape: [2, 1, 4, 1, 3], dataType: 'float32'}
}
},
'operators': [{
'name': 'sqrt',
'arguments': [{'input': 'sqrtInput'}],
'outputs': 'sqrtOutput'
}],
'expectedOutputs': {
'sqrtOutput': {
'data': [
2.693697690963745, 2.790419816970825, 1.1698113679885864,
2.8378541469573975, 3.0218021869659424, 2.919893980026245,
2.20693302154541, 2.7615139484405518, 1.644696593284607,
2.9512035846710205, 1.8079805374145508, 2.170917510986328,
2.944347381591797, 2.9714653491973877, 0.7435740828514099,
2.821129322052002, 2.023958206176758, 2.8142287731170654,
2.1034460067749023, 0.31454887986183167, 2.165646553039551,
1.2360169887542725, 2.4351139068603516, 2.4630208015441895
],
'descriptor': {shape: [2, 1, 4, 1, 3], dataType: 'float32'}
}
}
}
}
];
if (navigator.ml) {
sqrtTests.forEach((test) => {
webnn_conformance_test(
buildAndExecuteGraph, getSqrtPrecisionTolerance, test);
});
} else {
test(() => assert_implements(navigator.ml, 'missing navigator.ml'));
}