Credit Card Fraud Detection
Credit-Card-Fraud-Detection Overview Credit Card Fraud Detection project using a dataset from Kaggle, which includes 284,807 credit card transactions with only 492 frauds, is highly unbalanced. To address the class imbalance, the project implements a voting classifier and a neural network with focal loss in PyTorch, achieving an F1-score of 0.86 and 0.85 PR_AUC for the positive class (fraud). Installation Install the required packages 1 pip3 install -r requirements.txt Repo structure and File descriptions 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 credit Card Fraud Detection/ ├── config/ │ ├── config.yml │ └── trainer_config.yml │ ├── data/ │ ├── train.csv │ ├── val.csv │ └── test.csv.zip ├── models/ │ ├── │ └── ├── runs/ │ └── ... ├── .gitignore ├── cradit_fraud_utils_helper.py ├── credit_fraud_train.py ├── credit_fraud_utils_data.py ├── credit_fraud_utils_eval.py ├── eda.ipynb ├── focal_loss.py ├── model_evaluation.py ├── requirements.txt └── README.md config/ ...