Skip to content

Notebook for creating an ensemble classifier and tuning hyperparameters of underlying models

Notifications You must be signed in to change notification settings

feeney92/Ensemble_Classifier_Software_Defect_Prediction

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 

Repository files navigation

Ensemble_method_software_defect_prediction

In this notebook I explore how to create an ensemble classifier and how to hypertune the underlying models. The ensemble uses a variety of tree-based models (random forest, extra random trees, XGBoost, HistBoost, CatBoost, LightGBM) as well as a neural network. The model also uses a hill-climbing approach for determining the weights of the ensemble.

Whilst I have carried out some exploratory data analysis this is not the main focus of this notebook and should be explored further at a later date.

The overall aim of the model is to try to predict whether a piece of software will have any defects based on certain characteristics.

About

Notebook for creating an ensemble classifier and tuning hyperparameters of underlying models

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published