GitHub - DanTastic-Dev/Fraud-Detection: Applying the concepts of ML, we have developed a hybrid layer architecture for credit card fraud detection. · GitHub

Name		Name	Last commit message	Last commit date
layer1		layer1
layer2		layer2
layer3		layer3
.gitignore		.gitignore
README.md		README.md
finalstats.py		finalstats.py
finguard.py		finguard.py
informaldocs.txt		informaldocs.txt
requirements.txt		requirements.txt

Repository files navigation

FinGuard

Layer 1 — Done

Run: python3 layer1/main.py
Output: outputs/layer1_model.pkl, layer1_misses.csv, layer2_candidates.csv

Layer 2 — In Progress

Input: outputs/layer1_misses.csv + outputs/layer1_model.pkl
Task: RL environment that catches what Layer 1 missed
See layer2/ folder

Dataset

Download creditcard.csv from Kaggle and place in datasets/
Not pushed to repo (too large)

About

Applying the concepts of ML, we have developed a hybrid layer architecture for credit card fraud detection.

0 stars

0 watching

Report repository

Releases

Packages

Contributors

Languages