Evolutionary Automated Feature Engineering

The codes for paper Evolutionary Automated Feature Engineering

Configurations

First, preprocessing the dataset to a whole .csv file. Then, setting the feature information according to the dataset. Other configurations and hyperparameters are set in the main .py file, please reference to the examples.

How to run EAAFE code?

1. A sample run of the code is shown below

python test_eaafe.py

2. A sample run of the code, with generating transformed data

python test_eaafe_transform.py

3. A sample run of the code, with Ray

python test_eaafe_ray.py

4. A sample run of the code, with spark

python test_eaafe_spark.py

License

The codes and models in this repo are released under the GNU GPLv3 license.