— Interactive

Tools
& Labs

7 labs Ensembles · 4 Hyperparameters · 3

Live, in-browser machine-learning playgrounds — decision-boundary visualizers and hyperparameter sandboxes I built to make model behavior tangible. Tune the knobs and watch the boundary move.

/ 01 Ensembles

Voting Regressor Visualizer

Watch a voting ensemble blend several regressors into one smoother prediction surface — swap the base models and see the averaged fit update live.

Streamlit scikit-learn
Live demo Launch
/ 02 Ensembles

Voting Classifier Boundary Lab

See how hard and soft voting across multiple classifiers reshapes the decision boundary on 2D data in real time.

Streamlit scikit-learn
Live demo Launch
/ 03 Hyperparameters

Logistic Regression Sandbox

Tune regularization strength, penalty and solver for logistic regression and watch the decision boundary and metrics respond instantly.

Streamlit scikit-learn
Live demo Launch
/ 04 Hyperparameters

Decision Tree Sandbox

Sweep max depth, min-samples and split criteria on a decision-tree classifier to see over- and under-fitting emerge on the boundary.

Streamlit scikit-learn
Live demo Launch
/ 05 Hyperparameters

Decision Tree Regressor Sandbox

Explore how depth and leaf constraints turn a regression tree from a smooth step function into a jagged overfit — all interactively.

Streamlit scikit-learn
Live demo Launch
/ 06 Ensembles

Bagging Regressor Boundary Lab

A bagging-ensemble regressor lab showing how bootstrap aggregation reduces variance across many weak regressors.

Streamlit scikit-learn
Source only GitHub
/ 07 Ensembles

Bagging Classifier Boundary Lab

A bagging-ensemble classifier lab visualizing how aggregating bootstrapped models smooths a noisy decision boundary.

Streamlit scikit-learn
Source only GitHub