A lightweight and modular parallel PyTorch implementation of VICReg (intended for audio, but will try to be general)
Project description
vicregaddon
"vicreg add-on". As in "Yeah, but are you getting it? vicregaddon it"
A lightweight and modular parallel PyTorch implementation of VICReg.
(I intend this for audio, but could be used for other things)
Installation:
pip install vicregaddon
or, for the most recent version
pip install git+https://github.com/drscotthawley/vicregaddon.git
Usage:
from vicregaddon import *
Dependencies:
- torch
- einops
In the library itself, parallelism is handled directly via torch.distributed and thus is "agnostic" to whether PyTorch Lightning or Accelerate (or some other thing) is used.
The examples/ however, have several more dependencies. You will need to install Lightning and/or Accelerate and/or other things to run them, depending on each example.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file vicregaddon-0.0.4.tar.gz.
File metadata
- Download URL: vicregaddon-0.0.4.tar.gz
- Upload date:
- Size: 6.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.0.0 CPython/3.11.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
8d9f16b607ad5b27af605965384e61f216302ee567bc8e8db3d1cd803e14a8a0
|
|
| MD5 |
bf63ee5f0a418668bb68ce1b3ac4697c
|
|
| BLAKE2b-256 |
10a4fa87ea6d1eb4cf4334963919aea21acac8062a3eab4213bd8729ee71705f
|
File details
Details for the file vicregaddon-0.0.4-py3-none-any.whl.
File metadata
- Download URL: vicregaddon-0.0.4-py3-none-any.whl
- Upload date:
- Size: 7.4 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.0.0 CPython/3.11.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
419f2b04d2043ac6d3c71d29a09eecece24fa05d6feab9d3011ec5341347e3b3
|
|
| MD5 |
97bea42ae00f8fe6ebf606bf3fdaeae4
|
|
| BLAKE2b-256 |
32cb6c13a6b9a7f38dedc4e5a5c84539dcd1f49e99e91a0deb1bc2fe56dfe7a1
|