Machine Learning Library from Scratch with Python

👋 Introduction

In this project, I will "re-built from scratch" state-of-the-art algorithms in ML/DL, but using only Python language. I will use NumPy library for better performance on matrix calculation.

One thing you need to know that, I make this project just for learning and understanding algorithms deeply, not suitable to apply in real-world problems. I still recommend using many other SOTA libraries for building models.

👤 Contributors

I would like to express my sincere thanks to the these wonderful people who have contributed to this library with me:

• Duong T. (owner)
• Phuc P. (contributor)
• Thinh T. (contributor)

📝 Libraries

A. Classification

1. Logistic Regression | Document, Code
2. Naive Bayes| Document, Code
3. K-Nearest Neighbors (KNN) | Document, Code
4. Decision Tree | Document, Code
5. Support Vector Machine (SVM) | Document, Code
6. Random Forest
7. Softmax Regression

B. Regression

1. Linear Regression | Document, Code
2. Ridge Regression
3. Lasso Regression
4. Decision Tree for Regression
5. Random Forest for Regression
6. K-Nearest Neighbors for Regression
7. Support Vector Regression
8. Gaussian Regression
9. Polynomial Regression

C. Clustering

1. K-Means | Document, Code
2. DBSCAN
3. Mean Shift
4. OPTICS
5. Spectral Clustering
6. Mixture of Gaussians
7. Affinity Propagation
8. Agglomerative Clustering
9. BIRCH

D. Dimensionality reduction

1. Principal Components Analysis (PCA) | Document, Code
2. Factor Analysis (FA) | Document, Code
3. Linear Discriminant Analysis (LDA) | Document, Code
4. Truncated SVD
5. Kernel PCA
6. t-Distributed Stochastic Neighbor Embedding (t-SNE)
7. Multidimensional Scaling (MDS)
8. Isomap