Add ml_dtypes prefix for CC includes. Build #351: Commit 889f2c0 pushed by copybara-service bot main January 30, 2025 23:54 26m 36s Add tests of scalar type behavior Build #350: Commit ad2580b pushed by copybara-service bot main January 29, 2025 00:53 17m 38s Bump pypa...
"numpy>=2.1.0; python_version>='3.13'", ] [project.urls] homepage = "https://github.com/jax-ml/ml_dtypes" repository = "https://github.com/jax-ml/ml_dtypes" # Other: `documentation`, `changelog` [project.optional-dependencies] # Development deps (unittest, linting, formating,...) ...
ml_dtypesis a stand-alone implementation of several NumPy dtype extensions used in machine learning libraries, including: bfloat16: an alternative to the standardfloat16format 8-bit floating point representations, parameterized by number of exponent and mantissa bits, as well as the bias (if any)...
A stand-alone implementation of several NumPy dtype extensions used in machine learning. - Release v0.5.0 release · jax-ml/ml_dtypes
<remote-id type="github">jax-ml/ml_dtypes</remote-id> <remote-id type="pypi">ml_dtypes</remote-id> <remote-id type="pypi">ml-dtypes</remote-id> </upstream> </pkgmetadata>3 changes: 2 additions & 1 deletion 3 dev-python/ml_dtypes/ml_dtypes-0.2.0.ebuild → dev-python/ml-dtypes...
.nox/test_onnx_weekly/lib/python3.11/site-packages/onnx/reference/ops/_op.py:91: in run res = self._run(x, y) .nox/test_onnx_weekly/lib/python3.11/site-packages/onnx/reference/ops/_op.py:139: in _run res = (convert_from_ml_dtypes(res[0]),) .nox/test_onnx_weekly/lib/...
Actions: jax-ml/ml_dtypes Actions All workflows Build CodeQL Test Test with nightly NumPy Management Caches Attestations All workflows Showing runs from all workflows 556 workflow runs Event Status Branch Actor Scheduled CodeQL #189: by github-advanced-security bot July 26, ...
"numpy>=1.26.0; python_version>='3.12'", ] [project.urls] homepage = "https://github.com/jax-ml/ml_dtypes" repository = "https://github.com/jax-ml/ml_dtypes" # Other: `documentation`, `changelog` [project.optional-dependencies] # Development deps (unittest, linting, formating,......
A stand-alone implementation of several NumPy dtype extensions used in machine learning. - Add float8_e3m4 · jax-ml/ml_dtypes@4a03c71
"numpy>=1.26.0; python_version>='3.12'", ] [project.urls] homepage = "https://github.com/jax-ml/ml_dtypes" repository = "https://github.com/jax-ml/ml_dtypes" # Other: `documentation`, `changelog` [project.optional-dependencies] # Development deps (unittest, linting, formating,...)...