This repo contains Python implementations of fundamental machine learning classification algorithms. All algorithms heve been coded from scratch (no libraries) for a deep understanding of how each of them works. The algorithms include:
- Linear Regression Classifier with ridge regularization
- Gaussian naive Bayes Classifier
- Logistic Regression
- kernel Logistic Regression
- Neural Network Classifier with one or two hidden layer(s)
"script_classify.py" is the main script which instantiates and compares the performances of the classifiers. "classalgorithms.py" contains the actual implementations of the algorithms with each of them having their own separate class. Functions used to load the required datasets are available in "dataloader.py", and "utilities.py" provides other useful functions needed throughout the codes.