Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/paganpasta/eqxvision
A Python package of computer vision models for the Equinox ecosystem.
https://github.com/paganpasta/eqxvision
equinox python pytorch vision
Last synced: 6 days ago
JSON representation
A Python package of computer vision models for the Equinox ecosystem.
- Host: GitHub
- URL: https://github.com/paganpasta/eqxvision
- Owner: paganpasta
- License: mit
- Created: 2022-07-24T22:02:02.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-07-19T20:33:28.000Z (4 months ago)
- Last Synced: 2024-10-02T10:37:40.023Z (about 1 month ago)
- Topics: equinox, python, pytorch, vision
- Language: Python
- Homepage: https://eqxvision.readthedocs.io
- Size: 15.7 MB
- Stars: 100
- Watchers: 4
- Forks: 12
- Open Issues: 8
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
- awesome-jax - Eqxvision - Equinox version of Torchvision. <img src="https://img.shields.io/github/stars/paganpasta/eqxvision?style=social" align="center"> (Libraries / New Libraries)
README
# Eqxvision
[](https://pypi.org/project/eqxvision/)
[](https://eqxvision.readthedocs.io/en/latest/)
[](https://github.com/paganpasta/eqxvision/releases)
[](https://github.com/paganpasta/eqxvision/blob/main/LICENSE.md)
Eqxvision is a package of popular computer vision model architectures built using [Equinox](https://docs.kidger.site/equinox/).
## Installation
Use the package manager [pip](https://pip.pypa.io/en/stable/) to install eqxvision.
```bash
pip install eqxvision
```
*requires:* `python>=3.7`
*optional:* `torch`, only if `pretrained` models are required.
## Documentation
Available at [https://eqxvision.readthedocs.io/en/latest/](https://eqxvision.readthedocs.io/en/latest/).
## Usage
Picking a model and doing a forward pass is as simple as ...
```python
import jax
import jax.random as jr
import equinox as eqx
from eqxvision.models import alexnet
from eqxvision.utils import CLASSIFICATION_URLS
@eqx.filter_jit
def forward(net, images, key):
keys = jax.random.split(key, images.shape[0])
output = jax.vmap(net, axis_name=('batch'))(images, key=keys)
...
net = alexnet(torch_weights=CLASSIFICATION_URLS['alexnet'])
images = jr.uniform(jr.PRNGKey(0), shape=(1,3,224,224))
output = forward(net, images, jr.PRNGKey(0))
```
## What's New?
- `FCN`, `DeepLabV3` and `LRASPP` added as new image segmentation models.
- Backward incompatible changes to `v0.2.0` for loading a `pretrained` model.
- Almost all image classification models are ported from `torchvision`.
- New tutorial for generating `adversarial examples` and others coming soon.
## Get Started!
Start with any one of these easy to follow [tutorials](https://eqxvision.readthedocs.io/en/latest/getting_started/Transfer_Learning/).
## Tips
- Better to use `@equinox.filter_jit` instead of `@jax.jit`.
- Use `jax.{v,p}map` with `axis_name='batch'` when using models that use batch normalisation.
- Don't forget to switch to `inference` mode for evaluations. (`model = eqx.tree_inference(model)`)
- Initialise Optax optimisers as `optim.init(eqx.filter(net, eqx.is_array))`. ([See here.](https://docs.kidger.site/equinox/faq/#optax-is-throwing-an-error))
## Contributing
Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.
### Development Process
If you plan to modify the code or documentation, please follow the steps below:
1. Fork the repository and create your branch from `dev`.
2. If you have modified the code (new feature or bug-fix), please add unit tests.
3. If you have changed APIs, update the documentation. Make sure the documentation builds. `mkdocs serve`
4. Ensure the test suite passes. `pytest tests -vvv`
5. Make sure your code passes the formatting checks. Automatically checked with a `pre-commit` hook.
## Acknowledgements
- [Equinox](https://github.com/patrick-kidger/equinox)
- [Patrick Kidger](https://github.com/patrick-kidger)
- [Torchvision](https://pytorch.org/vision/stable/index.html)
## License
[MIT](https://choosealicense.com/licenses/mit/)