gfloat: Generic floating-point types in Python
An implementation of generic floating point encode/decode logic, handling various current and proposed floating point types:
- IEEE 754: Binary16, Binary32
- OCP Float8: E5M2, E4M3
- IEEE WG P3109: P3109_{K}p{P} for K > 2, and 1 <= P < K.
- OCP MX Formats: E2M1, M2M3, E3M2, E8M0, INT8, and the MX block formats.
The library favours readability and extensibility over speed (although the *_ndarray functions are reasonably fast for large arrays, see the benchmarking notebook). For other implementations of these datatypes more focused on speed see, for example, ml_dtypes, bitstring, MX PyTorch Emulation Library.
See https://gfloat.readthedocs.io for documentation, or dive into the notebooks to explore the formats.
For example, here's a table from the 02-value-stats notebook:
| name | B: Bits in the format | P: Precision in bits | E: Exponent field width in bits | Exact in float16? | Exact in float32? | 0<x<1 | 1<x<Inf | minSubnormal | maxSubnormal | minNormal | maxNormal | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| name | B | P | E | rt16 | rt32 | lt1 | gt1 | minSubnormal | maxSubnormal | minNormal | maxNormal | |
| -------------- | ----- | ----- | ----- | -------- | -------- | ------- | ------- | ---------------- | ---------------- | ------------- | --------------- | |
| p3109_k3p2sf | 3 | 2 | 1 | True | True | 1 | 1 | 0.5 | 0.5 | 1 | 1.5 | |
| ocp_e2m1 | 4 | 2 | 2 | True | True | 1 | 5 | 0.5 | 0.5 | 1 | 6 | |
| p3109_k4p2sf | 4 | 2 | 2 | True | True | 3 | 3 | 0.25 | 0.25 | 0.5 | 3 | |
| ocp_e2m3 | 6 | 4 | 2 | True | True | 7 | 23 | 0.125 | 0.875 | 1 | 7.5 | |
| ocp_e3m2 | 6 | 3 | 3 | True | True | 11 | 19 | 0.0625 | 0.1875 | 0.25 | 28 | |
| p3109_k6p3sf | 6 | 3 | 3 | True | True | 15 | 15 | 0.03125 | 0.09375 | 0.125 | 14 | |
| p3109_k6p4sf | 6 | 4 | 2 | True | True | 15 | 15 | 0.0625 | 0.4375 | 0.5 | 3.75 | |
| ocp_e4m3 | 8 | 4 | 4 | True | True | 55 | 70 | 2^-9 | 7/4*2^-7 | 0.015625 | 448 | |
| ocp_e5m2 | 8 | 3 | 5 | True | True | 59 | 63 | 2^-16 | 3/2*2^-15 | 2^-14 | 57344 | |
| p3109_k8p1se | 8 | 1 | 7 | False | True | 63 | 62 | n/a | n/a | 2^-63 | 2^62 | |
| p3109_k8p1ue | 8 | 1 | 8 | False | True | 127 | 125 | n/a | n/a | 2^-127 | 2^125 | |
| p3109_k8p3se | 8 | 3 | 5 | True | True | 63 | 62 | 2^-17 | 3/2*2^-16 | 2^-15 | 49152 | |
| p3109_k8p3sf | 8 | 3 | 5 | True | True | 63 | 63 | 2^-17 | 3/2*2^-16 | 2^-15 | 57344 | |
| p3109_k8p3ue | 8 | 3 | 6 | False | True | 127 | 125 | 2^-33 | 3/2*2^-32 | 2^-31 | 5/4*2^31 | |
| p3109_k8p3uf | 8 | 3 | 6 | False | True | 127 | 126 | 2^-33 | 3/2*2^-32 | 2^-31 | 3/2*2^31 | |
| p3109_k8p4se | 8 | 4 | 4 | True | True | 63 | 62 | 2^-10 | 7/4*2^-8 | 0.0078125 | 224 | |
| p3109_k8p4sf | 8 | 4 | 4 | True | True | 63 | 63 | 2^-10 | 7/4*2^-8 | 0.0078125 | 240 | |
| p3109_k8p4ue | 8 | 4 | 5 | True | True | 127 | 125 | 2^-18 | 7/4*2^-16 | 2^-15 | 53248 | |
| p3109_k8p4uf | 8 | 4 | 5 | True | True | 127 | 126 | 2^-18 | 7/4*2^-16 | 2^-15 | 57344 | |
| p3109_k8p7sf | 8 | 7 | 1 | True | True | 63 | 63 | 0.015625 | 63/32*2^-1 | 1 | 127/64*2^0 | |
| p3109_k8p8uf | 8 | 8 | 1 | True | True | 127 | 126 | 0.0078125 | 127/64*2^-1 | 1 | 127/64*2^0 | |
| binary16 | 16 | 11 | 5 | True | True | 15359 | 16383 | 2^-24 | 1023/512*2^-15 | 2^-14 | 65504 | |
| bfloat16 | 16 | 8 | 8 | False | True | 16255 | 16383 | 2^-133 | 127/64*2^-127 | 2^-126 | 255/128*2^127 | |
| ocp_e8m0 | 8 | 1 | 8 | False | True | 127 | 127 | n/a | n/a | 2^-127 | 2^127 | |
| ocp_int8 | 8 | 8 | 0 | True | True | 63 | 63 | 0.015625 | 127/64*2^0 | n/a | n/a |
Notes
All NaNs are the same, with no distinction between signalling or quiet, or between differently encoded NaNs.
1 maintainer
Metadata
1 maintainer
Release files for gfloat 0.5.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| gfloat-0.5.2.tar.gz | 657.9 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| gfloat-0.5.2-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 679.8 kB
Release files / gfloat-0.5.2.tar.gz
| Download URL | gfloat-0.5.2.tar.gz |
|---|---|
| Size | 657.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
51b11fc69a68b459d9f1cc1f03e33e10e30980428422b97cef2b5d851d116490
|
|
BLAKE2b-256 checksum How to use checksums |
37bea267e474b4182159e789a5dd1971c43d040924600cc97f058c23da3dc291
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.12.3
|
Release files / gfloat-0.5.2-py3-none-any.whl
| Download URL | gfloat-0.5.2-py3-none-any.whl |
|---|---|
| Size | 21.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
aaa2aceaffd7b282efa96ed8e4f1db0d7eef21e6d7683fe15d0dde234872e232
|
|
BLAKE2b-256 checksum How to use checksums |
1c1c952a3c5c4fdfcd7d240c317d1d7b2e37dcc58cf6adcd8c002fe1476ee942
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.12.3
|
