swvanderlaan/imbdrlappendix

The appendix for the paper on imbDRL

imbDRLAppendix

The appendix for the paper on Imbalanced Classification with Deep Reinforcement Learning: imbDRL.

Requirements

• Python 3.7+
• pip install -r requirements.txt
• Logs are by default saved in ./logs/
• Trained models are by default saved in ./models/
• Optional: ./data/ folder located at the root of this repository.
  • This folder must contain creditcard.csv downloaded from Kaggle if you would like to use the Credit Card Fraud dataset.
  • Note: creditcard.csv needs to be split in a seperate train and test file. Please use the function imbDRL.utils.split_csv

Running the experiments

• For the creditcard-fraud dataset:
  • Run ./experiments/creditcardfraud/nn_dta.py to run experiments for the standard NN and the DTA-method.
  • Run ./experiments/creditcardfraud/dqn.py to run experiments for DQN-algorithm.

Data for the histology and AKI datasets are not publicly available. The code for the experiments can be found in the ./experiments/histology and ./experiments/aki folders.