Update app.py

app.py

@@ -16,36 +16,48 @@ with open('category_encodings.pkl', 'rb') as f:
     category_encodings = pickle.load(f)
 
 # ---------------------------
-# App UI
+# Define column names (excluding 'attack_cat')
+# ---------------------------
+columns = [
+    'srcip', 'sport', 'dstip', 'dsport', 'proto', 'state', 'dur', 'sbytes', 'dbytes',
+    'sttl', 'dttl', 'sloss', 'dloss', 'service', 'Sload', 'Dload', 'Spkts', 'Dpkts',
+    'swin', 'dwin', 'stcpb', 'dtcpb', 'smeansz', 'dmeansz', 'trans_depth', 'res_bdy_len',
+    'Sjit', 'Djit', 'Stime', 'Ltime', 'Sintpkt', 'Dintpkt', 'tcprtt', 'synack', 'ackdat',
+    'is_sm_ips_ports', 'ct_state_ttl', 'ct_flw_http_mthd', 'is_ftp_login', 'ct_ftp_cmd',
+    'ct_srv_src', 'ct_srv_dst', 'ct_dst_ltm', 'ct_src_ ltm', 'ct_src_dport_ltm',
+    'ct_dst_sport_ltm', 'ct_dst_src_ltm', 'Label', 'duration', 'byte_ratio', 'pkt_ratio'
+]
+
+# ---------------------------
+# Streamlit UI
 # ---------------------------
 st.title("🚨 Intrusion Detection In Networks")
-st.markdown("Paste a single row of preprocessed NB15 features (comma-separated) below to predict the intrusion type.")
+st.markdown("""
+Paste a single preprocessed row of NB15 features (comma-separated) below.  
+Make sure it includes exactly **{}** comma-separated values matching the model's input format.
+""".format(len(columns)))
 
-user_input = st.text_area("Input (1 row from NB15_combined_preprocessed.csv):", height=150)
+user_input = st.text_area("📥 Input a row:", height=150)
 
 if st.button("Predict"):
     try:
-        # Convert user input string into a DataFrame
         row = [x.strip() for x in user_input.strip().split(",")]
-        input_df = pd.DataFrame([row])
-
-        # Load a sample row to get correct column names and data types
-        sample_df = pd.read_csv('NB15_combined_preprocessed.csv', nrows=1)
-        columns = sample_df.drop('attack_cat', axis=1).columns.tolist()
-
-        # Ensure correct number of columns
+        
         if len(row) != len(columns):
-            st.error(f"Expected {len(columns)} values, but got {len(row)}.")
+            st.error(f"❌ Expected {len(columns)} values, but got {len(row)}.")
         else:
-            input_df.columns = columns
-            # Convert to appropriate types
+            # Convert to DataFrame
+            input_df = pd.DataFrame([row], columns=columns)
+            
+            # Convert to numeric
             input_df = input_df.apply(pd.to_numeric, errors='coerce')
 
-            # Drop high-correlation features
+            # Drop the high-correlation features
             input_df = input_df.drop(columns=features_to_drop, errors='ignore')
 
-            # Predict
+            # Prediction
             prediction = model.predict(input_df)[0]
             st.success(f"🛡️ Predicted Intrusion Category: **{prediction}**")
+
     except Exception as e:
-        st.error(f"⚠️ Error processing input: {e}")
+        st.error(f"⚠️ Error during prediction: {e}")