Pre-release
This release is a pre-release and may not be stable for production use.
ml_dtypes
This is not an officially supported Google product.
ml_dtypes is a stand-alone implementation of several NumPy dtype extensions used in machine learning libraries, including:
bfloat16: an alternative to the standardfloat16formatfloat8_*: several experimental 8-bit floating point representations including:float8_e4m3b11float8_e4m3fnfloat8_e5m2
Installation
We will soon release pre-built wheels for this package at http://pypi.org/project/ml-dtypes, so this package can be easily pip installed.
To build locally, clone the repository and run:
$ git submodule init
$ git submodule update
$ pip install .
Example Usage
>>> from ml_dtypes import bfloat16
>>> import numpy as np
>>> np.zeros(4, dtype=bfloat16)
array([0, 0, 0, 0], dtype=bfloat16)
Importing ml_dtypes also registers the data types with numpy, so that they may
be referred to by their string name:
>>> np.dtype('bfloat16')
dtype(bfloat16)
>>> np.dtype('float8_e5m2')
dtype(float8_e5m2)
2 maintainers
Metadata
2 maintainers
Release files for ml-dtypes 0.0.2rc4
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| ml_dtypes-0.0.2rc4.tar.gz | 682.9 kB | Details |
Built distributions (wheels)
Total release size: 3.4 MB
Release files / ml_dtypes-0.0.2rc4.tar.gz
| Download URL | ml_dtypes-0.0.2rc4.tar.gz |
|---|---|
| Size | 682.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
2cfefe779e06ffaa330ae1d1e2ac1fbf3cef92206d0bd821fe4e651904bd8bed
|
|
BLAKE2b-256 checksum How to use checksums |
70d62207f173eeeae2adedffddbd46dda2f310f2bae86463f5ca75b6796036ef
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / ml_dtypes-0.0.2rc4-cp311-cp311-win_amd64.whl
| Download URL | ml_dtypes-0.0.2rc4-cp311-cp311-win_amd64.whl |
|---|---|
| Size | 135.4 kB |
| Tags | CPython 3.11 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
01529520c5382abb9abfcb5ea0fb57529f9b43bd2b7fc279078de60f5c26f3cf
|
|
BLAKE2b-256 checksum How to use checksums |
5337735ea2edf931790c86bfd7d00e7e3d7e5c17727cacef69c093659e9a9aaf
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / ml_dtypes-0.0.2rc4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
| Download URL | ml_dtypes-0.0.2rc4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl |
|---|---|
| Size | 231.0 kB |
| Tags | CPython 3.11 Linux glibc 2.17+ x86-64 |
|
SHA-256 checksum How to use checksums |
212009a3f90e729a6a2852986549482425c3d5e271d9f5892ed5cbe8271bc9df
|
|
BLAKE2b-256 checksum How to use checksums |
625f16ce60b790a3c63215d50222a8b8ebf3772f8e3631076320661f422824d8
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / ml_dtypes-0.0.2rc4-cp311-cp311-macosx_10_9_universal2.whl
| Download URL | ml_dtypes-0.0.2rc4-cp311-cp311-macosx_10_9_universal2.whl |
|---|---|
| Size | 312.0 kB |
| Tags | CPython 3.11 macOS 10.9+ universal2 (ARM64, x86-64) |
|
SHA-256 checksum How to use checksums |
b4d1d9863cea3794e7b6c46a9c67c12fef4f357398921e581722b8994f62290b
|
|
BLAKE2b-256 checksum How to use checksums |
081cf712baee9c938a4f95ac8ab4f3cbc9aa3c3a4fb00a1d49531764b1d34368
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / ml_dtypes-0.0.2rc4-cp310-cp310-win_amd64.whl
| Download URL | ml_dtypes-0.0.2rc4-cp310-cp310-win_amd64.whl |
|---|---|
| Size | 135.5 kB |
| Tags | CPython 3.10 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
9246a5e42dfaa64d015d21d48101f503e63b335ba3f6f51973a3ef9d6bf9361a
|
|
BLAKE2b-256 checksum How to use checksums |
23d117dc1bb41836f6e4743e18cd7a4c2ce35ad50edaa8f7fd901bd685669d47
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / ml_dtypes-0.0.2rc4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
| Download URL | ml_dtypes-0.0.2rc4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl |
|---|---|
| Size | 231.0 kB |
| Tags | CPython 3.10 Linux glibc 2.17+ x86-64 |
|
SHA-256 checksum How to use checksums |
517c4496007a200766f625f9a693f67c0e3ad28b79d54ab67f4b6ec5b6b3278c
|
|
BLAKE2b-256 checksum How to use checksums |
4cb9687ec6525a0616175255a4a43f24c38431aaf0ba3f7c579298dd07ba33dd
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / ml_dtypes-0.0.2rc4-cp310-cp310-macosx_10_9_universal2.whl
| Download URL | ml_dtypes-0.0.2rc4-cp310-cp310-macosx_10_9_universal2.whl |
|---|---|
| Size | 312.0 kB |
| Tags | CPython 3.10 macOS 10.9+ universal2 (ARM64, x86-64) |
|
SHA-256 checksum How to use checksums |
d12e05ff0ce313d4141ed49124a09d7a216d21957751adbf08f1054adce13d0a
|
|
BLAKE2b-256 checksum How to use checksums |
910ee2fa1b233205ca51b69154fc95ff381a09872c3653036cf30e607f1f0d2e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / ml_dtypes-0.0.2rc4-cp39-cp39-win_amd64.whl
| Download URL | ml_dtypes-0.0.2rc4-cp39-cp39-win_amd64.whl |
|---|---|
| Size | 135.4 kB |
| Tags | CPython 3.9 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
a7b10caf83c652be1bb6d24f1a87ba69a433d23a98115df9a528d9a117257d7a
|
|
BLAKE2b-256 checksum How to use checksums |
449389343aca4e62a4d4c03256d9e167a719e3bfe15f501e88346b95c17b78f9
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / ml_dtypes-0.0.2rc4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
| Download URL | ml_dtypes-0.0.2rc4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl |
|---|---|
| Size | 231.1 kB |
| Tags | CPython 3.9 Linux glibc 2.17+ x86-64 |
|
SHA-256 checksum How to use checksums |
c4ef24f64d62a63253a71b9ba848b4a4f15d5147a94db854d7e831fe570d2e64
|
|
BLAKE2b-256 checksum How to use checksums |
7a2df075e12882dd460126ec0318a77f70b1ddcca110b4887a74778b22b0c03c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / ml_dtypes-0.0.2rc4-cp39-cp39-macosx_10_9_universal2.whl
| Download URL | ml_dtypes-0.0.2rc4-cp39-cp39-macosx_10_9_universal2.whl |
|---|---|
| Size | 312.5 kB |
| Tags | CPython 3.9 macOS 10.9+ universal2 (ARM64, x86-64) |
|
SHA-256 checksum How to use checksums |
f99616f3304aaaf02ebe9b6912e4adf16b0e3364f35cb9df592617331744f773
|
|
BLAKE2b-256 checksum How to use checksums |
861f414656b9a29b094e4cbd1545b8a773e6eee99374188bb3570564d5c2ef05
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / ml_dtypes-0.0.2rc4-cp38-cp38-win_amd64.whl
| Download URL | ml_dtypes-0.0.2rc4-cp38-cp38-win_amd64.whl |
|---|---|
| Size | 134.8 kB |
| Tags | CPython 3.8 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
59666a3f94d6831210e8e8ca97fa8948ba099854f3548ffabe52304eddc4c423
|
|
BLAKE2b-256 checksum How to use checksums |
574224710448676c0f6861d26bb7dfef273a99e2fc0fe937ea1def0218bb6653
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / ml_dtypes-0.0.2rc4-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
| Download URL | ml_dtypes-0.0.2rc4-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl |
|---|---|
| Size | 231.0 kB |
| Tags | CPython 3.8 Linux glibc 2.17+ x86-64 |
|
SHA-256 checksum How to use checksums |
299763c225f14e3f0ea9a01d5b40df556406d8be993aa4ea72aa4423347dc6f1
|
|
BLAKE2b-256 checksum How to use checksums |
a5f693946b8198bdc8b04b40f4a13f954f2757e4786f7b29b70960e41edba2c6
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
Release files / ml_dtypes-0.0.2rc4-cp38-cp38-macosx_10_9_universal2.whl
| Download URL | ml_dtypes-0.0.2rc4-cp38-cp38-macosx_10_9_universal2.whl |
|---|---|
| Size | 311.9 kB |
| Tags | CPython 3.8 macOS 10.9+ universal2 (ARM64, x86-64) |
|
SHA-256 checksum How to use checksums |
c2080ad4822d1c139a3bd8385e71f7930c7207b2b56bc69cbc2307efff80045b
|
|
BLAKE2b-256 checksum How to use checksums |
6cf291296d0113d393be9496ef025cb1fd283bd2b907abf24619361d36958e59
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.16
|
