https://github.com/ae-bii/neural-geometry
Latent Space Geometry for Neural Networks in Python
https://github.com/ae-bii/neural-geometry
deep-learning machine-learning manifold-optimization neural-network python
Last synced: 7 months ago
JSON representation
Latent Space Geometry for Neural Networks in Python
- Host: GitHub
- URL: https://github.com/ae-bii/neural-geometry
- Owner: ae-bii
- License: mit
- Created: 2024-04-10T00:48:40.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2024-07-18T02:47:38.000Z (almost 2 years ago)
- Last Synced: 2025-09-28T18:23:55.499Z (9 months ago)
- Topics: deep-learning, machine-learning, manifold-optimization, neural-network, python
- Language: Python
- Homepage: https://ae-bii.github.io/neural-geometry/
- Size: 27.7 MB
- Stars: 17
- Watchers: 1
- Forks: 1
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
Awesome Lists containing this project
README
--------------------------------------------------------------------------------
> [!WARNING]
> This package is still in its early stages. Updates may cause breaking changes.
Neural Geometry is a Python library designed to explore and manipulate the geometric properties of neural network latent spaces. It provides a set of tools and methods to understand the complex, high-dimensional spaces that neural networks operate in, inspired by recent approaches (e.g. Borde et al., [2023](https://arxiv.org/pdf/2309.04810.pdf)).
The primary features of Neural Geometry include:
- An implementation of the neural latent geometry search framework. This framework provides a unique approach to product manifold inference, which can be beneficial in various fields such as machine learning and data analysis.
- A selection of optimization methods to cater to different needs and requirements. These methods can be used to fine-tune the performance of the neural latent geometry search framework.
This package is designed to be compatible with popular scientific computing libraries such as NumPy and PyTorch, making it a versatile tool for researchers and developers working in these environments. Comprehensive documentation is available at [docs](https://ae-bii.github.io/neural-geometry/).
## Installation
To install Neural Geometry, you can use pip:
```bash
pip install neural-geometry
```
You can install optional packages for development or visualization using:
```bash
pip install .[dev,vis] # install from pyproject.toml
pip install neural-geometry[dev,vis] # install from pypi
```
## Usage
After installing, you can import the package and use it by following the [example](examples/example.py).
## Contributing
Contributions to Neural Geometry are welcome! To contribute:
1. Fork the repository.
2. Install the pre-commit hooks using `pre-commit install`.
3. Create a new branch for your changes.
4. Make your changes in your branch.
5. Submit a pull request.
Before submitting your pull request, please make sure your changes pass all tests.
## License
Neural Geometry is licensed under the MIT License. See the [LICENSE](LICENSE) file for more details.