12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152 |
- Image Processing For Machine Learning
- =====================================
- This is a package developed during a thesis project.
- Installation
- ------------
- .. code:: bash
- pip install ipfml
- How to use ?
- ------------
- To use, simply do :
- .. code:: python
- from PIL import Image
- from ipfml.processing import transform
- img = Image.open('path/to/image.png')
- s = transform.get_LAB_L_SVD_s(img)
- Modules
- -------
- This project contains modules.
- - **metrics** : *Metrics computation for model performance*
- - **utils** : *All utils functions developed for the package*
- - **exceptions** : *All customized exceptions*
- - **filters** : *Image filter module with convolution*
- - **iqa** : *Image quality assessments*
- - **processing** : *Image processing module*
- All these modules will be enhanced during development of the package.
- Documentation
- -------------
- For more information about package, documentation_ is available.
- .. _documentation: https://prise-3d.github.io/ipfml/
- Contributing
- ------------
- Please refer to the guidelines_ file if you want to contribute!
- .. _guidelines: https://github.com/prise-3d/ipfml/blob/master/CONTRIBUTING.md
|