modify features encryption

label_encoders.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6828590f38440abf21c24157ec2c7ab8e25bf1100a277918afd2f5ead03f6912
+size 7719889

main.py

@@ -8,6 +8,8 @@ from fastapi.middleware.cors import CORSMiddleware
 from pydantic import BaseModel
 import pickle
 import joblib
+import pandas as pd
+from sklearn.preprocessing import LabelEncoder
 from extraction_features import extract_features 
 
 app = FastAPI()
@@ -24,7 +26,13 @@ app.add_middleware(
 # model_file = open('logistic_regression_model.pkl', 'rb')
 # model = pickle.load(model_file, encoding='bytes')
 model = joblib.load('logistic_regression_model.pkl')
-
+label_encoders = joblib.load('label_encoders.pkl')
+# Columns used in the model
+selected_columns = [
+    'URL', 'URLLength', 'Domain', 'DomainLength', 'TLD', 
+    'CharContinuationRate', 'TLDLength', 'NoOfSubDomain', 
+    'DegitRatioInURL', 'SpacialCharRatioInURL', 'IsHTTPS'
+]
 
 class Msg(BaseModel):
     msg: str
@@ -73,7 +81,17 @@ async def predict(request: Request, requess: Req = Depends(form_req)):
     url = requess.url
 
     features = extract_features(str(url))
-    data = []
+    dataFrame_features = pd.DataFrame([features])
+
+    # Apply features encoding (convert everything into int64)
+    for column in selected_columns:
+        if column in label_encoders: 
+            encoder = label_encoders[column]
+            dataFrame_features[column] = encoder.transform(dataFrame_features[column])
+
+    data = dataFrame_features[selected_columns].values
+
+    # data = []
     # data.append(str(features['URL']))
     # data.extend(int(features['URLLength']))
     # data.extend(str(features['Domain']))
@@ -85,19 +103,17 @@ async def predict(request: Request, requess: Req = Depends(form_req)):
     # data.extend(float(features['DegitRatioInURL']))
     # data.extend(float(features['SpacialCharRatioInURL']))
     # data.extend(int(features['IsHTTPS']))
-    data.append(features['URL'])
-    data.append(features['URLLength'])
-    data.append(features['Domain'])
-    data.append(features['DomainLength'])
-    data.append(features['TLD'])
-    data.append(features['CharContinuationRate'])
-    data.append(features['TLDLength'])
-    data.append(features['NoOfSubDomain'])
-    data.append(features['DegitRatioInURL'])
-    data.append(features['SpacialCharRatioInURL'])
-    data.append(features['IsHTTPS'])
-    # Convert every feature into numeric value
-    data = [float(x) if isinstance(x, str) else x for x in data]
+    # data.append(features['URL'])
+    # data.append(features['URLLength'])
+    # data.append(features['Domain'])
+    # data.append(features['DomainLength'])
+    # data.append(features['TLD'])
+    # data.append(features['CharContinuationRate'])
+    # data.append(features['TLDLength'])
+    # data.append(features['NoOfSubDomain'])
+    # data.append(features['DegitRatioInURL'])
+    # data.append(features['SpacialCharRatioInURL'])
+    # data.append(features['IsHTTPS'])
 
     prediction = model.predict([data])
     output = prediction[0]