{"library":"haliax","type":"library","category":null,"description":"Haliax (version 1.3) provides named tensors for JAX, enhancing legibility and reducing common shape-related errors in deep learning models. It builds on JAX's power by allowing users to refer to tensor dimensions by name, simplifying complex operations like broadcasting, reduction, and concatenation. The library is actively developed with frequent minor releases and occasional major updates.","language":"python","status":"active","version":"1.3","tags":["jax","deep-learning","named-tensors","machine-learning"],"last_verified":"Tue May 26","install":[{"cmd":"pip install haliax","imports":["from haliax import NamedArray","from haliax import Axis","from haliax import product","import haliax.nn as nn","from haliax.nn import Linear"]},{"cmd":"pip install haliax jax[cuda12_pip] -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html","imports":[]}],"homepage":null,"github":"https://github.com/stanford-crfm/haliax","docs":null,"changelog":null,"pypi":"https://pypi.org/project/haliax/","npm":null,"openapi_spec":null,"status_page":null,"smithery":null,"compatibility":{"summary":{"python_range":"3.10–3.9","success_rate":40,"avg_install_s":13.6,"avg_import_s":null,"wheel_type":"wheel"},"url":"https://checklist.day/v1/registry/haliax/compatibility"}}