jaxlib 0.3.7 with cuda 11.2

[ngam]

Checklist

• [x] Used a personal fork of the feedstock to propose changes
• [x] Bumped the build number (if the version is unchanged)
• [ ] Reset the build number to 0 (if the version changed)
• [x] Re-rendered with the latest conda-smithy (Use the phrase code>@<space/conda-forge-admin, please rerender in a comment in this PR for automated rerendering)
• [x] Ensured the license file is being packaged.

[conda-forge-linter]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe) and found it was in an excellent condition.

[ngam]

@conda-forge-admin, please rerender

[ngam]

A confirmation: py37 build from here (downloaded and uploaded artifact):

~$ mamba list | grep jax
jax                       0.3.10             pyhd8ed1ab_0    conda-forge
jaxlib                    0.3.7           cuda112py37h592705d_1    ngam

~$ python3 -c "import jax; import jaxlib; from jax.lib import xla_bridge; print(xla_bridge.get_backend().platform); print(jaxlib.__version__); print(jax.__version__)"
gpu
0.3.7
0.3.10

confirmation 2: from the artifact here building in less than 4 hours:

mamba create -n test39 python=3.9 jax jaxlib==*=*cuda112py39h75bc476* -c ngam -c conda-forge

all clear.

[ngam]

@conda-forge-admin, please rerender

[ngam]

@xhochy pls don't bother with the other PR. Please only review this, thanks! Only errors are connections dropping and I confirmed this works fine locally for 37, 38, 39, and 310

Update: the other PR targeting 0.3.10 is also ready. Tested locally and is good to go. Up to you if you want to release both 0.3.7 and 0.3.10, probably 0.3.10 is enough, but I will leave that up to you here. xref: #103

[ngam]

@conda-forge-admin, please rerender

[github-actions[bot]]

Hi! This is the friendly automated conda-forge-webservice.

I tried to rerender for you, but it looks like there was nothing to do.

This message was generated by GitHub actions workflow run https://github.com/conda-forge/jaxlib-feedstock/actions/runs/2341537524.

[ngam]

@conda-forge-admin, please rerender

[xhochy]

@ngam Can you switch this to using lief_dev as we did with tensorflow? I guess that could bring the builds faster over the finish line.

[ngam]

@ngam Can you switch this to using lief_dev as we did with tensorflow? I guess that could bring the builds faster over the finish line.

Good suggestion! I will do that soon and rerender

[ngam]

@conda-forge-admin, please rerender

[ngam]

Not really sure why some never work here randomly. The version of this PR for jaxlib 0.3.10 (#103), things seem to finish better...