Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

`compile_with_cache()` is mistakenly exposed to public?

See original GitHub issue

I don’t know how it started (I am sure it’s not in CuPy’s doc at all) but I’ve seen several bug reports from the ML community calling this function (such as #5294) https://github.com/cupy/cupy/blob/4469fae998df33c72ff40ef954cb08b8f0004b18/cupy/cuda/__init__.py#L95 Since we already have cupy.RawModule to meet this need (and beyond), I think we can safely remove it from cupy/cuda/__init__.py, perhaps with a deprecation warning.

Issue Analytics

State:
Created 2 years ago
Comments:5 (5 by maintainers)

Top GitHub Comments

2reactions

kmaehashicommented, May 31, 2021

RawKernel introduced in v5 and RawModule introduced in v7. compile_with_cache has never been documented API, but it’s widely used by major users like Thinc and 3d-ken-burns which need Raw* features before we implement it.

I basically agree with the deprecation but we also need to take care of existing users. I think we need some more steps than usual deprecations, e.g.: