Update app.py

app.py

@@ -1,61 +1,43 @@
-import pandas as pd
 import pickle
+import pandas as pd
 import streamlit as st
-# Load the trained model from data.pkl (assuming it's a custom model without sklearn)
-def load_model():
-  with open('data.pkl', 'rb') as file:
-    model = pickle.load(file)
-  return model
-
-# Define the prediction function using the loaded model
-def predict_user_profile(inputs):
-  # Preprocess the input data (assuming your model doesn't require LabelEncoder)
-  # ... (modify this section based on your specific model's preprocessing needs)
-
-  # Create a DataFrame from the user input dictionary
-  df = pd.DataFrame.from_dict([inputs])
-
-  # Select the relevant feature columns used during model training
-  feature_columns_to_use = ['statuses_count', 'followers_count', 'friends_count',
-                            'favourites_count','listed_count']  # Assuming language isn't used
-  df_features = df[feature_columns_to_use]
-
-  # Load the pre-trained model
-  model = load_model()
-
-  # Make predictions using the loaded model (assuming your model has a predict method)
-  prediction = model.predict(df_features)
 
-  # Return the predicted class label (modify based on your model's output)
-  if prediction > 0.5:  # Assuming a threshold for genuine classification
-    return "Genuine"
-  else:
-    return "Fake"
-
-# Create the Streamlit app
-st.title('User Profile Classifier')
-st.write('Predict whether a user profile is genuine or fake.')
-
-# Create input fields for user data
-statuses_count = st.number_input("Statuses Count", min_value=0)
-followers_count = st.number_input("Followers Count", min_value=0)
-friends_count = st.number_input("Friends Count", min_value=0)
-favourites_count = st.number_input("Favourites Count", min_value=0)
-listed_count = st.number_input("Listed Count", min_value=0)
-name = st.text_input("Name")
-# Remove language input since it's not used in this version
-
-# Create a dictionary to store user inputs
-user_input = {
-  "statuses_count": statuses_count,
-  "followers_count": followers_count,
-  "friends_count": friends_count,
-  "favourites_count": favourites_count,
-  "listed_count": listed_count,
-  "name": name,
+# Load the trained model
+model = pickle.load(open("data.pkl", "rb"))
+
+# Define a function to predict user data
+def predict_user_data(user_data):
+  user_df = pd.DataFrame(user_data, index=[0])
+  user_df = extract_features(user_df)  # Assuming the extract_features function is defined elsewhere in your code
+  prediction = model.predict(user_df)[0]
+  return prediction
+
+# Streamlit app layout
+st.title("Fake or Genuine User Classifier")
+
+# Get user input
+user_statuses_count = st.number_input("Statuses Count", min_value=0)
+user_followers_count = st.number_input("Followers Count", min_value=0)
+user_friends_count = st.number_input("Friends Count", min_value=0)
+user_favourites_count = st.number_input("Favourites Count", min_value=0)
+user_listed_count = st.number_input("Listed Count", min_value=0)
+user_name = st.text_input("Name")
+
+# Get user input as a dictionary
+user_data = {
+    "statuses_count": user_statuses_count,
+    "followers_count": user_followers_count,
+    "friends_count": user_friends_count,
+    "favourites_count": user_favourites_count,
+    "listed_count": user_listed_count,
+    "name": user_name,
 }
 
 # Predict if the user clicks the button
-if st.button("Predict"):
-  prediction = predict_user_profile(user_input)
-  st.write(f"Prediction: {prediction}")
+if st.button("Classify User"):
+  prediction = predict_user_data(user_data)
+  if prediction == 1:
+    st.success("The user is likely Genuine.")
+  else:
+    st.warning("The user is likely Fake.")
+