GitHub - ypeleg/HungaBunga: HungaBunga: Brute-Force all sklearn models with all parameters using .fit .predict!

Abstraction: Python library brute-forcing all sklearn models and hyperparameters via cross-validation

Key points:

• HungaBunga runs through all scikit-learn supervised learning models (classification and regression) with all possible hyperparameter combinations, ranked by cross-validation
• Covers Generalized Linear Models, SVMs, Nearest Neighbors, Gaussian Processes, Naive Bayes, Trees, Neural Networks, and Ensemble methods
• Usage mirrors standard sklearn API: HungaBungaClassifier().fit(x, y).predict(x)
• Motivated by the view that most supervised ML work is feature engineering, not model selection — this automates the latter
• In example output, ExtraTreesClassifier achieved top score of 0.993 vs SGDClassifier at 0.967 and NuSVC at 0.980

Connections: Scikit Learn · Automated Machine Learning · Model Selection · Hyperparameter Search

Source: https://github.com/ypeleg/HungaBunga