A collection of independent machine learning notebooks focused on core algorithms, model behavior, evaluation problems, and from-scratch implementations.
The repository includes technical studies on decision trees, boosting, PCA, support vector machines, bias-variance trade-off, target leakage, model benchmarking, and Kaggle-style modeling workflows.
The notebooks are written in a simple and direct style, with an emphasis on clear reasoning, structured experimentation, and practical interpretation of results.
This repository is designed as a technical machine learning portfolio rather than a general tutorial series.
Each notebook focuses on a specific machine learning concept, algorithm, or modeling problem.
Some notebooks rebuild algorithms from scratch, while others investigate how models behave, how evaluation can become misleading, and how different modeling choices affect the final result.
The notebooks focus on:
- from-scratch machine learning implementations
- model behavior and evaluation
- bias, variance, and leakage
- decision trees, boosting, PCA, and SVM
Note: Additional notebooks and technical studies will continue to be added over time.