Skip to content
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.

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

Update all dpnp tests to run on IRIS Xe #1293

Closed
vlad-perevezentsev opened this issue Feb 13, 2023 · 1 comment
Closed

Update all dpnp tests to run on IRIS Xe #1293

vlad-perevezentsev opened this issue Feb 13, 2023 · 1 comment
Assignees

Comments

@vlad-perevezentsev
Copy link
Collaborator

Most tests fail when running them on gpu device without double type support (Iris XE).
Need to update all dpnp tests to consider the specifics of this device.

test_absolute.py output example from WSL:

=short test summary info =
FAILED tests/test_absolute.py::test_absolute_float[float64] - AssertionError: 
FAILED tests/test_absolute.py::test_absolute_float_3[float64] - AssertionError: 
=2 failed, 2 passed in 0.38s =

E   AssertionError: 
E   Arrays are not equal
E   
E   Mismatched elements: 1 / 9 (11.1%)
E   Max absolute difference: 3.81469727e-07
E   Max relative difference: 4.19197485e-08
E    x: array([[2. , 3. , 9.1],
E          [2. , 5. , 2. ],
E          [1. , 2. , 5. ]])
E    y: array([[2. , 3. , 9.1],
E          [2. , 5. , 2. ],
E          [1. , 2. , 5. ]], dtype=float32)
@vlad-perevezentsev
Copy link
Collaborator Author

At the moment most of dpnp tests have been fixed to run on Iris Xe.
The remaining tests will be fixed as soon as all functions from dpctl are reused.
You can find all failed tests in tests/skipped_tests_gpu_no_fp64.tbl

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

1 participant