Update app.py

app.py

@@ -4,55 +4,110 @@ import joblib
 import numpy as np
 import plotly.graph_objects as go
 import os
+import xgboost as xgb
 
-# Page Config
-st.set_page_config(page_title="Jet Engine AI", page_icon="✈️")
+# 1. PAGE SETUP
+st.set_page_config(page_title="Jet Engine AI Predictor", page_icon="✈️", layout="wide")
+
+# 2. AUTO-TRAINING LOGIC (Fixes the "Missing File" or "Corrupted File" issue)
+def train_and_save_model():
+    with st.spinner("Training the AI model for the first time... please wait."):
+        # Load Data from NASA Repository
+        url = "https://raw.githubusercontent.com/datasets-machine-learning/nasa-turbofan-failure-prediction/master/data/train_FD001.txt"
+        cols = ['unit', 'cycles', 'os1', 'os2', 'os3'] + [f's{i}' for i in range(1, 22)]
+        df = pd.read_csv(url, sep='\s+', header=None, names=cols)
+
+        # Calculate RUL (Target)
+        max_cycles = df.groupby('unit')['cycles'].max().reset_index()
+        max_cycles.columns = ['unit', 'max_of_unit']
+        df = df.merge(max_cycles, on='unit', how='left')
+        df['RUL'] = df['max_of_unit'] - df['cycles']
+
+        # Select Features (15 columns)
+        features = ['cycles', 's2', 's3', 's4', 's7', 's8', 's11', 's12', 's13', 's15', 's17', 's20', 's21', 'os1', 'os2']
+        X = df[features]
+        y = df['RUL']
+
+        # Train XGBoost
+        model = xgb.XGBRegressor(n_estimators=100, learning_rate=0.1, max_depth=5)
+        model.fit(X, y)
+        
+        # Save locally in the HF Space
+        joblib.dump(model, 'engine_model.pkl')
+        return model
 
 @st.cache_resource
 def load_model():
     if os.path.exists('engine_model.pkl'):
-        return joblib.load('engine_model.pkl')
-    return None
+        try:
+            return joblib.load('engine_model.pkl')
+        except:
+            # If the file is corrupted (like the mport error), re-train it
+            return train_and_save_model()
+    else:
+        return train_and_save_model()
 
+# Load model on startup
 model = load_model()
 
-st.title("✈️ Predictive Maintenance Dashboard")
-
-if model is None:
-    st.error("Model file 'engine_model.pkl' not found. Please upload it.")
-    st.stop()
-
-# Sidebar for Inputs
-st.sidebar.header("Sensor Data Input")
-cycle = st.sidebar.slider("Operational Cycle", 1, 350, 100)
-s2 = st.sidebar.number_input("Sensor 2 (Temp)", value=642.0)
-s3 = st.sidebar.number_input("Sensor 3 (Temp)", value=1589.0)
-s4 = st.sidebar.number_input("Sensor 4 (Pressure)", value=1408.0)
-s7 = st.sidebar.number_input("Sensor 7 (HPC Press)", value=553.0)
-s11 = st.sidebar.number_input("Sensor 11 (Speed)", value=47.5)
-
-if st.button("Predict Engine Health"):
-    # Create input array (matches the 15 features in training)
-    # Filling remaining slots with defaults for simplicity
-    inputs = np.array([[cycle, s2, s3, s4, s7, 550, s11, 2388, 521, 8.4, 392, 39, 23, 0.001, 0.0002]])
-    prediction = model.predict(inputs)
-    rul = int(max(0, prediction[0]))
-
-    # Gauge Chart
-    fig = go.Figure(go.Indicator(
-        mode = "gauge+number",
-        value = rul,
-        title = {'text': "Remaining Useful Life (Cycles)"},
-        gauge = {'axis': {'range': [0, 200]},
-                 'steps' : [
-                     {'range': [0, 30], 'color': "red"},
-                     {'range': [30, 80], 'color': "orange"},
-                     {'range': [80, 200], 'color': "green"}]}))
-    st.plotly_chart(fig)
+# 3. UI LAYOUT
+st.title("✈️ Jet Engine Predictive Maintenance System")
+st.markdown("---")
+
+col1, col2 = st.columns([1, 2])
+
+with col1:
+    st.header("📥 Sensor Inputs")
+    st.write("Adjust sliders to simulate engine conditions:")
     
-    if rul < 30:
-        st.error("🚨 CRITICAL: Maintenance Required Immediately!")
-    elif rul < 80:
-        st.warning("⚠️ CAUTION: Schedule Maintenance Soon.")
+    # Inputs
+    cycle = st.slider("Current Cycles", 1, 350, 100)
+    s2 = st.slider("Temp (S2)", 640.0, 650.0, 642.5)
+    s3 = st.slider("Temp (S3)", 1580.0, 1600.0, 1589.0)
+    s4 = st.slider("Pressure (S4)", 1400.0, 1430.0, 1408.0)
+    s7 = st.slider("Pressure (S7)", 550.0, 560.0, 553.5)
+    s11 = st.slider("Speed (S11)", 47.0, 48.5, 47.5)
+    
+    # Analyze Button
+    predict_btn = st.button("Predict Engine Health", type="primary")
+
+with col2:
+    st.header("📊 Prediction Result")
+    
+    if predict_btn:
+        # Prepare inputs (must be 15 values to match model training)
+        # Using some static defaults for the other 9 features to keep UI simple
+        sample_input = np.array([[cycle, s2, s3, s4, s7, 550, s11, 2388, 521, 8.4, 392, 39, 23, 0.001, 0.0002]])
+        
+        # Make Prediction
+        prediction = model.predict(sample_input)
+        rul = int(max(0, prediction[0]))
+
+        # Gauge Chart
+        fig = go.Figure(go.Indicator(
+            mode = "gauge+number",
+            value = rul,
+            title = {'text': "Remaining Useful Life (Cycles)"},
+            gauge = {
+                'axis': {'range': [0, 200]},
+                'steps' : [
+                    {'range': [0, 30], 'color': "red"},
+                    {'range': [30, 80], 'color': "orange"},
+                    {'range': [80, 200], 'color': "green"}],
+                'bar': {'color': "black"}
+            }
+        ))
+        st.plotly_chart(fig)
+
+        # Status Message
+        if rul < 30:
+            st.error(f"🚨 CRITICAL: Engine failure likely in {rul} cycles!")
+        elif rul < 80:
+            st.warning(f"⚠️ CAUTION: Maintenance check required within {rul} cycles.")
+        else:
+            st.success(f"✅ HEALTHY: Engine is safe for {rul} more cycles.")
     else:
-        st.success("✅ HEALTHY: Engine is in good condition.")
+        st.info("Adjust the sliders and click 'Predict' to see the results.")
+
+st.markdown("---")
+st.caption("B.Tech AI&DS Project | Dataset: NASA C-MAPSS")