refactor(prediction): refactor the controller code

src/agents/l1_screener.py

@@ -0,0 +1,49 @@
+import joblib
+from fastapi import HTTPException
+
+class Screener:
+    def __init__(self, model_path: str = "src/models/L1_Logistic_v0.joblib"):
+        try:
+            saved = joblib.load(model_path)
+            self.model = saved["model"]
+            self.scaler = saved["scaler"]
+            self.features = [
+                "Header_Length",
+                "Time_To_Live",
+                "Rate",
+                "Tot sum",
+                "Tot size",
+                "Min",
+                "Max",
+                "AVG",
+                "Std",
+                "Variance",
+                "IAT",
+                "Number"
+            ]
+            self.pca = saved.get("pca", None)
+        except FileNotFoundError:
+            raise HTTPException(status_code=500, detail=f"Modèle '{model_path}' introuvable")
+        except Exception as e:
+            raise HTTPException(status_code=500, detail=f"Erreur lors du chargement du modèle: {e}")
+
+    def predict(self, data):
+        try:
+            # Préparation des features
+            if "label" in data.columns:
+                X = data[self.features]
+            else:
+                X = data[self.features]
+
+            # Standardisation
+            X = self.scaler.transform(X)
+
+            # PCA si utilisé
+            if self.pca is not None:
+                X = self.pca.transform(X)
+
+            # Prédictions
+            preds = self.model.predict(X)
+            return preds.tolist()
+        except Exception as e:
+            raise HTTPException(status_code=500, detail=f"Erreur lors de la prédiction: {e}")

src/controllers/prediction_controller.py

@@ -1,133 +1,31 @@
-import pandas as pd
-import numpy as np
-import joblib
-import io
 from fastapi import HTTPException, UploadFile
-
-# Colonnes requises
-REQUIRED_COLUMNS = [
-    "Header_Length", "Protocol Type", "Time_To_Live", "Rate",
-    "fin_flag_number", "syn_flag_number", "rst_flag_number",
-    "psh_flag_number", "ack_flag_number", "ece_flag_number",
-    "cwr_flag_number", "ack_count", "syn_count", "fin_count",
-    "rst_count", "TCP", "UDP", "Tot sum", "Min", "Max", "AVG",
-    "Std", "Tot size", "IAT", "Number", "Variance"
-]
+from src.services.preprocess_features import preprocess_features
+from src.services.file_verification import file_verification
+from src.agents.l1_screener import Screener
 
 def global_prediction_on_csv(file: UploadFile):
     try:
-        # Vérifier l'extension
-        print("start verifying file extension")
-        if not file.filename.endswith(".csv"):
-            raise HTTPException(status_code=400, detail="Le fichier doit être un CSV")
-        
-        print("Correct file extension")
-
-        # Lecture du CSV depuis le contenu binaire
-        contents = file.file.read()
-        data = pd.read_csv(io.BytesIO(contents))
-
-        print("start verifying required columns")
-        # Vérification des colonnes
-        missing_cols = [col for col in REQUIRED_COLUMNS if col not in data.columns]
-        if missing_cols:
-            raise HTTPException(
-                status_code=400,
-                detail=f"Colonnes manquantes dans le CSV: {', '.join(missing_cols)}"
-            )
-
-        print("start data preprocessing")
-        # Nettoyage
-        data = data.replace([np.inf, -np.inf], np.nan).dropna()
-
-        eps = 1e-6
-        number = data["Number"] + eps
-        iat = data["IAT"] + eps
-        min_val = data["Min"] + 1
-
-        # Ratios de flags
-        data["syn_ratio"] = data["syn_count"] / number
-        data["ack_ratio"] = data["ack_count"] / number
-        data["fin_ratio"] = data["fin_count"] / number
-        data["rst_ratio"] = data["rst_count"] / number
-
-        # Tailles de paquets
-        data["mean_pkt_size"] = data["Tot size"] / number
-        data["pkt_size_range"] = data["Max"] - data["Min"]
-        data["pkt_size_ratio"] = data["Max"] / min_val
+        # Vérifier l'extension et les colonnes du fichier
+        correct_csv = file_verification(file)
 
-        # Timing
-        data["mean_iat"] = data["IAT"] / number
-        data["pkt_rate"] = data["Number"] / iat
+        # augmenter le nombre de features
+        data = preprocess_features(correct_csv)
 
-        # Débit
-        data["throughput"] = data["Tot size"] / iat
-        data["bytes_per_sec"] = data["Rate"] * data["Tot size"]
-
-        # Variation
-        data["coef_var"] = data["Std"] / (data["AVG"] + eps)
-
-        # Indicateurs combinés
-        data["tcp_udp_ratio"] = data["TCP"] / (data["UDP"] + 1)
-        data["flag_entropy"] = (
-            data["syn_ratio"] + data["ack_ratio"] + data["fin_ratio"] + data["rst_ratio"]
-        )
-        print("finish data preprocessing")
 
         print("start loading model and predicting")
 
-        # Chargement du modèle
-        try:
-            saved = joblib.load("src/models/L1_Logistic_v0.joblib")
-            model = saved["model"]
-            scaler = saved["scaler"]
-            pca = saved.get("pca", None)
-            # features = saved["features"]
-        except FileNotFoundError:
-            raise HTTPException(status_code=500, detail="Modèle 'L1_Logistic_v0.joblib' introuvable")
-        # try:
-        #     model = joblib.load("src/models/L1_Logistic_v0.joblib")
-        # except FileNotFoundError:
-        #     raise HTTPException(status_code=500, detail="Modèle 'L1_Logistic_v0.joblib' introuvable")
-
-        print("drop label column if exists")
-
-        # Préparation des features
-        features = [
-            "Header_Length",
-            "Time_To_Live",
-            "Rate",
-            "Tot sum",
-            "Tot size",
-            "Min",
-            "Max",
-            "AVG",
-            "Std",
-            "Variance",
-            "IAT",
-            "Number"
-        ]
-
-        if "label" in data.columns:
-            X = data[features]
-        else:
-            X = data[features]
-            
-        print("features prepared, starting prediction")
-
-        # Standardisation
-        X = scaler.transform(X)
-
-        # PCA si utilisé
-        if pca is not None:
-            X = pca.transform(X)
-        # Prédictions
-        preds = model.predict(X)
+        screener = Screener()
+        predictions = screener.predict(data)
 
+        total_samples = len(predictions)
+        total_attacks = sum(predictions)           # 1 = attaque
+        total_normal = total_samples - total_attacks  # 0 = requête normale
         return {
             "message": "Prédictions effectuées avec succès",
-            "total_samples": len(preds),
-            "predictions": preds.tolist()
+            "total_samples": len(predictions),
+            "total_attacks": total_attacks,
+            "total_normal_requests": total_normal,
+            "predictions": predictions.tolist()
         }
 
     except HTTPException:

src/services/file_verification.py

@@ -0,0 +1,48 @@
+from fastapi import HTTPException, UploadFile
+import pandas as pd
+import io
+
+
+# Colonnes requises
+required_columns = [
+    "Header_Length", "Protocol Type", "Time_To_Live", "Rate",
+    "fin_flag_number", "syn_flag_number", "rst_flag_number",
+    "psh_flag_number", "ack_flag_number", "ece_flag_number",
+    "cwr_flag_number", "ack_count", "syn_count", "fin_count",
+    "rst_count", "TCP", "UDP", "Tot sum", "Min", "Max", "AVG",
+    "Std", "Tot size", "IAT", "Number", "Variance"
+]
+
+def file_verification(file: UploadFile) -> pd.DataFrame:
+    """
+    Vérifie l'extension et les colonnes du fichier CSV, puis retourne un DataFrame valide.
+    Lève une HTTPException descriptive en cas d'erreur.
+    """
+    try:
+        print("start verifying file extension")
+        if not file.filename.endswith(".csv"):
+            raise HTTPException(status_code=400, detail="Le fichier doit être un CSV")
+        
+        print("Correct file extension")
+
+        # Lecture du CSV depuis le contenu binaire
+        contents = file.file.read()
+        data = pd.read_csv(io.BytesIO(contents))
+
+        print("start verifying required columns")
+
+        # Vérification des colonnes
+        missing_cols = [col for col in required_columns if col not in data.columns]
+        if missing_cols:
+            raise HTTPException(
+                status_code=400,
+                detail=f"Colonnes manquantes dans le CSV: {', '.join(missing_cols)}"
+            )
+
+        print("file verification successful")
+        return data
+
+    except HTTPException:
+        raise
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Erreur de lecture du fichier: {e}")

src/services/preprocess_features.py

@@ -0,0 +1,48 @@
+import pandas as pd
+import numpy as np
+
+
+def preprocess_features(data: pd.DataFrame) -> pd.DataFrame:
+    """
+    Prépare et enrichit les features du CSV avant prédiction.
+    """
+    
+    print("start data preprocessing")
+    data = data.replace([np.inf, -np.inf], np.nan).dropna()
+
+    eps = 1e-6
+    number = data["Number"] + eps
+    iat = data["IAT"] + eps
+    min_val = data["Min"] + 1
+
+    # Ratios de flags
+    data["syn_ratio"] = data["syn_count"] / number
+    data["ack_ratio"] = data["ack_count"] / number
+    data["fin_ratio"] = data["fin_count"] / number
+    data["rst_ratio"] = data["rst_count"] / number
+
+    # Tailles de paquets
+    data["mean_pkt_size"] = data["Tot size"] / number
+    data["pkt_size_range"] = data["Max"] - data["Min"]
+    data["pkt_size_ratio"] = data["Max"] / min_val
+
+    # Timing
+    data["mean_iat"] = data["IAT"] / number
+    data["pkt_rate"] = data["Number"] / iat
+
+    # Débit
+    data["throughput"] = data["Tot size"] / iat
+    data["bytes_per_sec"] = data["Rate"] * data["Tot size"]
+
+    # Variation
+    data["coef_var"] = data["Std"] / (data["AVG"] + eps)
+
+    # Indicateurs combinés
+    data["tcp_udp_ratio"] = data["TCP"] / (data["UDP"] + 1)
+    data["flag_entropy"] = (
+        data["syn_ratio"] + data["ack_ratio"] + data["fin_ratio"] + data["rst_ratio"]
+    )
+    
+    print("finish data preprocessing")
+
+    return data