Update app.py

app.py

@@ -30,16 +30,16 @@ tags_metadata = [
 
 load_dotenv()
 
-# Variables MLflow : URI de tracking, nom du modèle et stage
+# Mlflow variables
 MLFLOW_TRACKING_APP_URI = os.getenv("MLFLOW_TRACKING_APP_URI")
 MODEL_NAME = os.getenv("MODEL_NAME", "fraud_detection_dtc")
 STAGE = os.getenv("STAGE", "production")
 
-# Variables AWS pour accéder au bucket S3 qui contient les artifacts de MLflow
+# AWS variables
 os.environ["AWS_ACCESS_KEY_ID"] = os.getenv("AWS_ACCESS_KEY_ID")
 os.environ["AWS_SECRET_ACCESS_KEY"] = os.getenv("AWS_SECRET_ACCESS_KEY")
 
-# Variables globales pour stocker le modèle
+# Load model
 mlflow.set_tracking_uri(MLFLOW_TRACKING_APP_URI)
 model_uri = f"models:/{MODEL_NAME}@{STAGE}"
 model = mlflow.sklearn.load_model(model_uri)
@@ -95,7 +95,7 @@ def predict(features: PredictionFeatures):
     Returns:
         dict: A dictionary containing the prediction of whether the payment is fraud or not
     """
-    df = pd.DataFrame({
+    data = pd.DataFrame({
         "category" : [features.category],
         "amt" : [features.amt],
         "merch_fraud_level" : [features.merch_fraud_level],
@@ -107,8 +107,7 @@ def predict(features: PredictionFeatures):
     })
     
     try:
-        df = pd.DataFrame({...})
-        return model.predict(df)[0]
+        return model.predict(data)[0]
     
     except Exception as e:
         return {"error": str(e)}