A compact notebook for comparing how different supervised classifiers behave under the same split and feature-preparation path.

Models explored

• Logistic Regression
• K-Nearest Neighbors
• Gaussian Naive Bayes
• Decision Tree
• Support Vector Machine with an RBF kernel

The notebook uses a train/test split, standardizes features where distance or margin geometry matters, and compares predictions with accuracy, confusion matrices, and classification reports.

Run

git clone https://github.com/ReaperXD67/Supervised-ML-Models.git
cd Supervised-ML-Models
python -m venv .venv
pip install -r requirements.txt
jupyter notebook Untitled.ipynb

Evaluation discipline

The purpose is model-behavior comparison, not a leaderboard claim. For a stronger study, wrap preprocessing and estimators in scikit-learn pipelines, use stratified cross-validation, tune only inside training folds, and report per-class precision/recall/F1 alongside accuracy.