/manigp

Manifold Learning using Genetic Programming

Primary LanguageJupyter NotebookMIT LicenseMIT

manigp

Manifold Learning with GP

How to run the software

The command needs to contain the word Classifer in ML method.

python evaluate_model.py -ml NameClassifier /source/to/dataset -save_file /source/to/result/file -seed number

An example launching command looks as follows: python evaluate_model.py -ml GPClassifier datasets.txt -save_file results.txt -seed 1393

The hyperparameters for each classifier are located in methods/NameOfClassifier.py. The file requires definition of hyper_params, as well as initialization of a classifier as est.

The tests were run on the datasets from PMLB repository: https://github.com/EpistasisLab/penn-ml-benchmarks/

forked from https://github.com/athril/manigp