QuentinJassey/epicoin-fraud-xgboost

Epicoin Fraud Detection - XGBoost Model

Modèle de détection de fraude pour transactions financières.

Métriques

• Recall: 99.7%
• Precision: 96.5%
• F1-Score: 98.1%
• Latency P95: 0.23ms

Dataset

Entraîné sur PaySim (6.3M transactions).

Usage

import requests

API_URL = "https://api-inference.huggingface.co/models/QuentinJassey/epicoin-fraud-xgboost"
headers = {"Authorization": "Bearer YOUR_HF_TOKEN"}

payload = {
    "inputs": {
        "amount": 5000,
        "oldbalanceOrg": 10000,
        "newbalanceOrig": 5000,
        "type_encoded": 1
    }
}

response = requests.post(API_URL, headers=headers, json=payload)
print(response.json())