A fast, yet specialized, RMSNorm/LayerNorm implementation
Reason this release was yanked:
Replaced by faster-norm.
Project description
# fast-norm-cuda
A fast, yet specialized, RMSNorm/LayerNorm implementation
This library is under development. Currently, only some special cases are supported, and the performance is not yet fully optimized.
[x] RMSNorm
[ ] LayerNorm
[x] Float16 and BFloat16
[ ] More data types
[x] More shapes
[ ] Accelerate if no wgrad
[ ] Performance tuning
## Statement
This work was independently completed by me at home using my personal RTX 3080.
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
fast_norm_cuda-0.2.0.tar.gz
(6.1 kB
view details)
File details
Details for the file fast_norm_cuda-0.2.0.tar.gz.
File metadata
- Download URL: fast_norm_cuda-0.2.0.tar.gz
- Upload date:
- Size: 6.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.0.0 CPython/3.11.8
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 398660a39f27b8fdb0c008a2773abde23aa669c0976c0b50d63a8fec300f2cb0 |
|
| MD5 | 740b2ff1ed0e28e45bd69031fefef939 |
|
| BLAKE2b-256 | 42c364eccb2f384d5f96338b937e607cb986d1acab923bfa8bae81f1795b0b22 |
