Closed priti-ashvin-shah-ibm closed 2 years ago
Information from @mtreinish! Thank you for the information.
There was a configuration format change in numpy made in their most recent release 1.22 (which they decided to release on Dec 31 and that happened to break how qutip was parsing the numpy config.
I don't need all of you to approve the pull request. Just wanted you to know this is the fix. As a sanity check, could one you use Metal with qutip and confirm all is behaving as desired?
In reflection, fixing this in tox wasn't logical since qutip is used in more than just testing.
When this is merged, issue #752 can be closed and branch can be deleted.
Should work fine, but couldn't finish testing due to EPR bug I ran into (though the two should be entirely unrelated)
… When fixed, we can remove the limit.
What are the issues this pull addresses (issue numbers / links)?
Qutip has an issue. https://github.com/qutip/qutip/pull/1752 is the pull request that will help. After this is available, we can remove the limit on numpy.
Did you add tests to cover your changes (yes/no)?
Did you update the documentation accordingly (yes/no)?
Did you read the CONTRIBUTING document (yes/no)?
Summary
Details and comments