HamBa-m/scinis-learn

Scinis-learn is a package of non-OOP functions for Machine Learning developed by young Moroccan AI engineering students from scratch. ## Overview In this package, we've implemented a few basic machine learning algorithms from scratch. The algorithms are implemented in Python 3.8. The algorithms are as follows: - Single Layer Perceptron Algorithm - Pocket Perceptron - Adaline Algorithm with delta rule - Linear Regression - Logistic Regression - Polynomial Regression - One-vs-All and One-vs-One Classifiers Furthermore, we've implemented a few basic algorithms useful in the cristalization of learning theory : - Non-Linear Transformation - Cross Validation with K-Fold - Gradient Descent for Linear and Logistic Regression - Regularization for Linear and Logistic Regression - Bias and Variance (not implemented yet) - Vapnik Chervonenkis Dimension (VC-Dimension) - Covering and Uniform Covering Number Finally, we also added a few basic functions to help you create dummy data and plot results of the algorithms : - Generate Dummy Data (in 2D and 3D) - Plot Linear Decision Boundary (in 2D and 2D) - Plot Non-Linear Decision Boundary (in 2D) - Plot Linear Regression (in 2D) All the algorithms and tools are implemented in the `lib` folder. These codes were made in the context of the Learning Theory course Fall 2023 at ENSIAS - University Mohammed V - Rabat, Morocco. You can find the LABS and the corresponding PDFs in the `ensias_labs` folder. Do not use them since they are a beta version with a lot of bugs and errors. ## Authors and Contributors - Hamza Bamohammed, Applied Mathematics & AI engineering student at ENSIAS - Hicham Filali, Applied Mathematics & AI engineering student at ENSIAS - Bouchra Sahri, Applied Mathematics & AI engineering student at ENSIAS - Mohammed Nechba, Applied Mathematics & AI engineering student at ENSIAS - Hanaa El Afia, Applied Mathematics & AI engineering student at ENSIAS - Mohamed Mouhajir, Applied Mathematics & AI engineering student at ENSIAS ## Installation To install the package, you can use the following command : ```bash git clone github.com …