Skip to content

Stacking Machine Learning Models. Tunning; feature engineering, scaling, models combinations and parameters.

Notifications You must be signed in to change notification settings

CodeByHarri/Stacking-Ensemble-Machine-Learning

Repository files navigation

Stacked Generalization or Ensemble Machine Learning

The following notebook is a work in progress.

This notebook aims to be a template for classification or regression tasks by stacking three machine learning models together. Using the hyper-optimisation library: Optuna, three of the above models are chosen, scaled uniquely and tunned. These three predictions are then tunned inside the above equation.

Models used in this notebook:

- Multi-layer Perceptron 
- K-Neighbours
- Support Vector
- Gaussian Process   +  Radial-Basis Function  
- Gaussian Naïve Bayes
- Quadratic Discriminant Analysis
- Linear Regression 
- Lasso
- Ridge
- Ada Boost
- Extra Trees
- Random Forest
- XGBoost
- LightGBM

Areas for possible improvement:

- Additional parameters to tune

About

Stacking Machine Learning Models. Tunning; feature engineering, scaling, models combinations and parameters.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published