HyperclassifierSearch

General info

HyperclassifierSearch allows to train multiple classifiers/pipelines in Python with GridSearchCV or RandomizedSearchCV.

Installation

pip install HyperclassifierSearch

Requirements

The code was developed in Python 3. The execution needs Pandas and scikit-learn, i.e. GridSearchCV and RandomizedSearchCV.

Enhancements and credits

The package is build based on code from David Batista.

1. documentation enhancements:

• examples how to search the best model over multiple Pipelines using different classifiers
• added code documentation including docstrings

2. functionality enhancements:

• added option to use RandomizedSearchCV
• the best overall model is provided by train_model()
• output dataframe is simplified as standard option

Examples

Please refer to HyperclassifierSearch examples.ipynb in the root folder.