Use virtualenv if needed.
$ virtualenv {working directory}
$ cd {working directory}
Start virtualenv.
$ source bin/activate
Or source activate.csh if you are using not bash.
Then install the required packages.
$ pip install -r requirements.txt
Install imbalanced-learn:
$ package-setup.sh
Just add new fields to config.ini.
Use 'get' method provided by configparser instead of directly field access so that the relationship with the configparser package will not be so strong.
All Sampler implementation should be placed in Sampler and should inherit BaseSampler. The name of the sampler should be added to SAMPLER.allowed in config.ini, and the corresponding strategy should be added to sampler.py. The name of the file should be added to __init__.py in Sampler.
All classifier implementation should be placed in Classifier and shoud inherit BaseClassifier. The name of the classifier should be added to CLASSIFIER.allowed in config.ini, and the corresponding stategy should be added to classifier.py. The name of the file should be added to __init__.py in Classifier.
All utility files should be placed in Utils. The name of the methods should be added to __init__.py.