jaxlib 0.9.0.1

XLA library for JAX

Stars: 34897, Watchers: 34897, Forks: 3425, Open Issues: 2208

The jax-ml/jax repo was created 7 years ago and the last code push was 9 minutes ago.
The project is extremely popular with a mindblowing 34897 github stars!

How to Install jaxlib

You can install jaxlib using pip

pip install jaxlib

or add it to a project with poetry

poetry add jaxlib

Package Details

Author

JAX team

License

Apache-2.0

Homepage

https://github.com/jax-ml/jax

PyPi:

https://pypi.org/project/jaxlib/

GitHub Repo:

https://github.com/google/jax

Classifiers

Related Packages

Errors

A list of common jaxlib errors.

Code Examples

Here are some jaxlib code examples and snippets.

GitHub Issues

The jaxlib package has 2208 open issues on GitHub

• It is not possible to define the .vma of FFI outputs
• Convenient interface for raising exceptions via a conditional host-callback
• [Doc] Update jnp.argmin/argmax doc for Pallas TPU
• Automated Code Change
• [Pallas TPU] jnp.argmin/argmax returns last index instead of first index on ties
• Automated Code Change
• Support device_put of an uncommitted array to a single global device in McJAX.
• [ROCm] Skip test_batch_axis_sharding_jvp on ROCm
• _pathways.pyi: Allow creating an XLA CPU client.
• [ROCm] Skip sparse tests on ROCm due to hipSPARSE issue
• Enable miscellaneous tests on ROCm
• Fix and enable Pallas ops tests on ROCm
• Enable neural network tests on ROCm
• Enable scaled_matmul tests on ROCm
• [Mosaic GPU] Generalize the fast i8 -> bf16 conversion to allow more vector lengths

See more issues on GitHub