google/jax

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

PythonC++StarlarkJupyter NotebookCShelljax
This is stars and forks stats for /google/jax repository. As of 04 May, 2024 this repository has 24923 stars and 2351 forks.

JAX: Autograd and XLA Quickstart | Transformations | Install guide | Neural net libraries | Change logs | Reference docs What is JAX? JAX is Autograd and XLA, brought together for high-performance numerical computing, including large-scale machine learning research. With its updated version of Autograd, JAX can automatically differentiate native Python and NumPy functions. It can differentiate through loops, branches, recursion, and closures, and it can take derivatives of derivatives of derivatives....
Read on GithubGithub Stats Page
repotechsstarsweeklyforksweekly
linkedin/FastTreeSHAPPythonJupyter NotebookC++4640290
elastic/detection-rulesPythonCC++1.6k04000
facebookresearch/pifuhdPythonOther9.2k01.4k0
AsahiLinux/asahi-installerPythonShell6300740
cyrus-and/gdb-dashboardPython9.7k07510
mne-tools/mne-pythonPythonCsound DocumentShell2.4k+51.3k+5
goauthentik/authentikPythonTypeScriptGo4.8k03760
pandas-dev/pandasPythonCythonHTML40k016.8k0
electricitymap/electricitymap-contribPythonHTMLTypeScript3.1k08530
allegroai/clearmlPython4.8k06200