You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,
I'm the lead developer of NetKet, an established machine learning / quantum physics package.
We have recently finished rewriting our core to be based on Jax (and flax), and recently released a beta version.
Since many physicists seem to use anaconda, we would also like to update our conda recipe.
However, since we depend on optax (and therefore on Chex), we would need Chex to have a Conda recipe.
Is that something you'd consider? I am willing to volunteer some work to help you.
I tried creating a recipe starting from your pypi source distribution, but that is problematic because you don't bundle your requirements.txt file, which is required to run setup.py.
I could create a recipe from the tag tarballs on GitHub, but that sometimes prevent the conda packages from auto-updating the recipe for later releases.
The text was updated successfully, but these errors were encountered:
Hi,
I'm the lead developer of NetKet, an established machine learning / quantum physics package.
We have recently finished rewriting our core to be based on Jax (and flax), and recently released a beta version.
Since many physicists seem to use anaconda, we would also like to update our conda recipe.
However, since we depend on optax (and therefore on Chex), we would need Chex to have a Conda recipe.
Is that something you'd consider? I am willing to volunteer some work to help you.
I tried creating a recipe starting from your pypi source distribution, but that is problematic because you don't bundle your
requirements.txt
file, which is required to runsetup.py
.I could create a recipe from the tag tarballs on GitHub, but that sometimes prevent the conda packages from auto-updating the recipe for later releases.
The text was updated successfully, but these errors were encountered: