Remove some hard-coded np.linalg
calls from tests
#501
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Outdated suggestions cannot be applied.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.
In this PR and in qiboteam/qibojit#36 I am trying to fix issue qiboteam/qibojit#34 concerning tests with AMD GPUs.
In particular, there are two tests in Qibo with calls to
np.linalg.eigh
andnp.linalg.eigvalsh
.When these calls are performed with a CuPy object as argument, it seems that somehow there is a automatic call to
cp.linalg.eigh
orcp.linalg.eigvalsh
(I think that this behavior is very confusing).The problem is that this behavior bypasses the fallback mechanism implemented in qiboteam/qibojit#36 and a
rocblas_not_implemented_status
error is raised.The easiest fix is to replace the hard-coded call to NumPy with the backend
K
in these tests.However, I wonder if there was a specific reason to use NumPy in these tests, e.g. to compare the results with
K
.@scarrazza @stavros11 what do you think?