An attempt to create and maintain a repository that contains a lot/most, if not all, machine learning data models.
The reason behind the creation of the repository was to develop and maintain a structured file of various, prevalent machine learning models to allow ease of re-usability of the code in future. The datasets used are the simplest I can find. The repositories' focus is solely on machine learning models, rather than preprocessing, evaluation, visualizatoins or data.
- Basic understanding and usage of all the Machine Learning models using sklearn.
- Working with fundamental pre-processing techniques in sklearn and visualizations using Matplotlib.
- Best evaluation practices and exposure to various metrics.
- Versatile yet simple choice of data
- Add more models
- Add cross validation for each model.
- Increase the volume of helpful/important/relevant links