Relax CI constraint on numpy<1.25
#11020
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The major problem in NumPy 1.25 transpired to be that particular versions of Apple clang (including the one used in Numpy's wheel-build CD pipelines) produced non-deterministic code from new SIMD routines for complex multiplication. This in turn destabilised our `Isometry` decompositions, which caused major correctness problems. As of Numpy 1.26.1 (released 2023-10-14), that SIMD code now decays to the scalar form if the compiler was known to be buggy, which restores the previous behaviour.
jakelishman
added
stable backport potential
|
One or more of the the following people are requested to review this:
|
|
The common errors between Linux/macOS/Windows and Numpy 1.26.1 are fixed by #11023. I'm not certain if the Linux+ |
Pull Request Test Coverage Report for Build 6619917150
💛 - Coveralls |
mtreinish
approved these changes
Oct 27, 2023
github-merge-queue
bot
removed this pull request from the merge queue due to failed status checks
mergify bot
pushed a commit
that referenced
this pull request
Nov 10, 2023
The major problem in NumPy 1.25 transpired to be that particular versions of Apple clang (including the one used in Numpy's wheel-build CD pipelines) produced non-deterministic code from new SIMD routines for complex multiplication. This in turn destabilised our `Isometry` decompositions, which caused major correctness problems. As of Numpy 1.26.1 (released 2023-10-14), that SIMD code now decays to the scalar form if the compiler was known to be buggy, which restores the previous behaviour. (cherry picked from commit fb8a69d)
mtreinish
added a commit
to mtreinish/qiskit-core
that referenced
this pull request
Nov 10, 2023
Removing the restriction on the numpy version has introduced some non-determinism to some of the unitary synthesis tests that is causing a flaky failure in CI. This commit reverts the version cap removal to unblock CI. In parallel we should debug the source of the instability causing the test to fail so we can run CI using the latest version of numpy. This reverts commit fb8a69d.
github-merge-queue bot
pushed a commit
that referenced
this pull request
Nov 11, 2023
Removing the restriction on the numpy version has introduced some non-determinism to some of the unitary synthesis tests that is causing a flaky failure in CI. This commit reverts the version cap removal to unblock CI. In parallel we should debug the source of the instability causing the test to fail so we can run CI using the latest version of numpy. This reverts commit fb8a69d.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
The major problem in NumPy 1.25 transpired to be that particular versions of Apple clang (including the one used in Numpy's wheel-build CD pipelines) produced non-deterministic code from new SIMD routines for complex multiplication. This in turn destabilised our
Isometrydecompositions, which caused major correctness problems. As of Numpy 1.26.1 (released 2023-10-14), that SIMD code now decays to the scalar form if the compiler was known to be buggy, which restores the previous behaviour.Details and comments
See numpy/numpy#24000 for details.
Fix #10305
Close #10906 (obsolete now that there's a fixed Numpy released)