Overview
A complete, self-contained ML reference built as Jupyter notebooks. Covers the stack from scratch: Python fundamentals, data preprocessing, classical ML, deep learning, and NLP. Each folder is a standalone topic with its own datasets and examples.
What It Covers
| Topic | Content |
|---|---|
| Python | Core language, data structures, OOP, standard library |
| Data Science | Missing values, encoding, scaling, train/test split |
| Regression | Linear, polynomial, SVR, random forest |
| Classification | Logistic regression, KNN, Naive Bayes, SVM, decision trees, model comparison |
| Clustering | K-means |
| Deep Learning | Neural networks, customer churn, audiobooks, MNIST |
| NLP | Text preprocessing, sentiment analysis, NER, LDA topic modeling, spaCy (10 notebooks) |
Why It Exists
Built as a personal reference while working through the ML curriculum — structured to be reproducible from scratch. Each notebook is self-contained: load the dataset, run the cells, see the output.