Skip to content

[core] Change inputs to matMul test to avoid precision issues in Safari as stopgap. #3142

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Apr 24, 2020
Merged

[core] Change inputs to matMul test to avoid precision issues in Safari as stopgap. #3142

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
a0e1c8e
add test
annxingyuan Apr 24, 2020
2424ee8
add issue link
annxingyuan Apr 24, 2020
d37ca40
Merge branch 'master' into fix_conv_flaky_Test
lina128 Apr 24, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
25 changes: 16 additions & 9 deletions tfjs-core/src/ops/conv2d_test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -328,18 +328,25 @@ describeWithFlags('conv2d', ALL_ENVS, () => {
const pad = 'same';
const stride: [number, number] = [2, 2];

const inputs = generateCaseInputs(
1 * xSize * xSize * inputDepth, fSize * fSize * inputDepth);
const x = tf.tensor4d(inputs.input, inputShape);
const w =
tf.tensor4d(inputs.filter, [fSize, fSize, inputDepth, outputDepth]);

// TODO(annxingyuan): Make this test work with large inputs using
// generateCaseInputs https://github.com/tensorflow/tfjs/issues/3143
const inputData = [];
for (let i = 0; i < xSize * xSize * inputDepth; i++) {
inputData.push(i % 5);
}

const wData = [];
for (let i = 0; i < fSize * fSize * inputDepth * outputDepth; i++) {
wData.push(i % 5);
}

const x = tf.tensor4d(inputData, inputShape);
const w = tf.tensor4d(wData, [fSize, fSize, inputDepth, outputDepth]);
const result = tf.conv2d(x, w, stride, pad);
expect(result.shape).toEqual([1, 4, 4, 1]);
expectArraysClose(await result.data(), new Float32Array([
2209560, 2543640, 2877720, 1890576, 4882200, 5216280,
5550360, 3475728, 7554840, 7888920, 8223000, 5060880,
4153744, 4302736, 4451728, 2551904
854, 431, 568, 382, 580, 427, 854, 288, 431, 568, 580,
289, 285, 570, 285, 258
]));
});

Expand Down