Update app.py

app.py

@@ -1,203 +1,208 @@
-import streamlit as st
-import pandas as pd
-import numpy as np
-import joblib
-import random
-from sklearn.preprocessing import StandardScaler
-from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, confusion_matrix
-from tensorflow.keras.models import load_model
-import warnings
-warnings.filterwarnings("ignore")
-
-# Health tips
-HEALTH_TIPS = [
-    "Regular exercise can reduce your risk of heart disease by up to 30%.",
-    "Eating a handful of nuts daily can improve heart health.",
-    "Reducing salt intake can help lower blood pressure.",
-    "Aim for 7-8 hours of quality sleep each night for heart health.",
-    "Quitting smoking can cut your heart disease risk in half within a year.",
-    "Managing stress through meditation can benefit your cardiovascular system."
-]
-
-# Load models
-def load_ml_model(model_path):
-    return joblib.load(model_path)
-
-def load_dl_model(model_path):
-    return load_model(model_path)
-
-# Load scaler
-scaler = load_ml_model("models/scaler.sav")
-
-# Page config
-st.set_page_config(page_title="CardioPredictor", page_icon="❤️")
-
-def main():
-    st.title("Cardiovascular Risk Predictor App ❤️")
-    
-    # Display random health tip
-    st.info(f"💡 Health Tip: {random.choice(HEALTH_TIPS)}")
-    
-    # Create input form
-    with st.form("prediction_form"):
-        st.subheader("Patient Information")
-        
-        col1, col2 = st.columns(2)
-        
-        with col1:
-            age = st.number_input("Age", min_value=20, max_value=100, value=50)
-            sex = st.selectbox("Sex", ["Male", "Female"])
-            cp = st.selectbox("Chest Pain Type", [
-                "Typical angina", 
-                "Atypical angina", 
-                "Non-anginal pain", 
-                "Asymptomatic"
-            ])
-            trestbps = st.number_input("Resting Blood Pressure (mm Hg)", min_value=80, max_value=200, value=120)
-            chol = st.number_input("Serum Cholesterol (mg/dl)", min_value=100, max_value=600, value=200)
-            fbs = st.selectbox("Fasting Blood Sugar > 120 mg/dl", ["No", "Yes"])
-        
-        with col2:
-            restecg = st.selectbox("Resting ECG Results", [
-                "Normal",
-                "ST-T wave abnormality",
-                "Left ventricular hypertrophy"
-            ])
-            thalach = st.number_input("Maximum Heart Rate Achieved", min_value=60, max_value=220, value=150)
-            exang = st.selectbox("Exercise Induced Angina", ["No", "Yes"])
-            oldpeak = st.number_input("ST Depression Induced by Exercise", min_value=0.0, max_value=6.2, value=1.0, step=0.1)
-            slope = st.selectbox("Slope of Peak Exercise ST Segment", [
-                "Upsloping",
-                "Flat",
-                "Downsloping"
-            ])
-            ca = st.selectbox("Number of Major Vessels Colored by Fluoroscopy", [0, 1, 2, 3])
-            thal = st.selectbox("Thalassemia", [
-                "Normal",
-                "Fixed defect",
-                "Reversible defect"
-            ])
-        
-        model_choice = st.selectbox("Select Prediction Model", [
-            "Logistic Regression",
-            "Artificial Neural Network (ANN)",
-            "Long Short-Term Memory (LSTM)",
-            "XGBoost"
-        ])
-        
-        submitted = st.form_submit_button("Predict")
-    
-    if submitted:
-        # Convert inputs to model format
-        sex = 1 if sex == "Male" else 0
-        cp_mapping = {
-            "Typical angina": 0,
-            "Atypical angina": 1,
-            "Non-anginal pain": 2,
-            "Asymptomatic": 3
-        }
-        cp = cp_mapping[cp]
-        
-        fbs = 1 if fbs == "Yes" else 0
-        
-        restecg_mapping = {
-            "Normal": 0,
-            "ST-T wave abnormality": 1,
-            "Left ventricular hypertrophy": 2
-        }
-        restecg = restecg_mapping[restecg]
-        
-        exang = 1 if exang == "Yes" else 0
-        
-        slope_mapping = {
-            "Upsloping": 0,
-            "Flat": 1,
-            "Downsloping": 2
-        }
-        slope = slope_mapping[slope]
-        
-        thal_mapping = {
-            "Normal": 1,
-            "Fixed defect": 2,
-            "Reversible defect": 3
-        }
-        thal = thal_mapping[thal]
-        
-        # Prepare input data
-        input_data = {
-            'age': age,
-            'sex': sex,
-            'cp': cp,
-            'trestbps': trestbps,
-            'chol': chol,
-            'fbs': fbs,
-            'restecg': restecg,
-            'thalach': thalach,
-            'exang': exang,
-            'oldpeak': oldpeak,
-            'slope': slope,
-            'ca': ca,
-            'thal': thal
-        }
-        
-        # Preprocess input
-        processed_input = scaler.transform(pd.DataFrame([input_data]))
-        
-        try:
-            # Load model and make prediction
-            if model_choice == "Logistic Regression":
-                model = load_ml_model("models/logistic_model.sav")
-                proba = model.predict_proba(processed_input)[0]
-                prediction = model.predict(processed_input)
-                
-            elif model_choice == "Artificial Neural Network (ANN)":
-                model = load_dl_model("models/ann_model.h5")
-                proba = model.predict(processed_input)[0][0]
-                proba = [1-proba, proba]  # Convert to [prob_0, prob_1]
-                prediction = [1 if proba[1] > 0.5 else 0]
-                
-            elif model_choice == "Long Short-Term Memory (LSTM)":
-                model = load_dl_model("models/lstm_model.h5")
-                lstm_input = processed_input.reshape(1, 1, processed_input.shape[1])
-                proba = model.predict(lstm_input)[0][0]
-                proba = [1-proba, proba]  # Convert to [prob_0, prob_1]
-                prediction = [1 if proba[1] > 0.5 else 0]
-                
-            elif model_choice == "XGBoost":
-                model = load_ml_model("models/xgb_model.sav")
-                proba = model.predict_proba(processed_input)[0]
-                prediction = model.predict(processed_input)
-
-            # 1. Probability Output
-            st.subheader("Prediction Results")
-            if prediction[0] == 1:
-                st.error(f"High Risk of Cardiovascular Disease (Probability: {proba[1]:.2%})")
-            else:
-                st.success(f"Low Risk of Cardiovascular Disease (Probability: {proba[0]:.2%})")
-
-            # 2. Evaluation Metrics
-            st.subheader("Model Performance Metrics")
-            test_data = pd.read_csv("heart.csv")
-            X_test = test_data.drop('target', axis=1)
-            y_test = test_data['target']
-            X_test_scaled = scaler.transform(X_test)
-            
-            if model_choice in ["Logistic Regression", "XGBoost"]:
-                y_pred = model.predict(X_test_scaled)
-            elif model_choice == "Artificial Neural Network (ANN)":
-                y_pred = (model.predict(X_test_scaled) > 0.5).astype(int).flatten()
-            elif model_choice == "Long Short-Term Memory (LSTM)":
-                lstm_test = X_test_scaled.reshape(X_test_scaled.shape[0], 1, X_test_scaled.shape[1])
-                y_pred = (model.predict(lstm_test) > 0.5).astype(int).flatten()
-            
-            # Calculate metrics
-            st.write(f"Accuracy: {accuracy_score(y_test, y_pred):.2%}")
-            st.write(f"Precision: {precision_score(y_test, y_pred):.2%}")
-            st.write(f"Recall: {recall_score(y_test, y_pred):.2%}")
-            st.write(f"F1 Score: {f1_score(y_test, y_pred):.2%}")
-
-        except Exception as e:
-            st.error(f"Error during prediction: {str(e)}")
-
-if __name__ == '__main__':
+import streamlit as st
+import pandas as pd
+import numpy as np
+import joblib
+import random
+from sklearn.preprocessing import StandardScaler
+from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, confusion_matrix
+import os
+os.environ["PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION"] = "python"  # 🔥 Must be set before TF import
+
+from tensorflow.keras.models import load_model
+
+#from tensorflow.keras.models import load_model
+import warnings
+warnings.filterwarnings("ignore")
+
+# Health tips
+HEALTH_TIPS = [
+    "Regular exercise can reduce your risk of heart disease by up to 30%.",
+    "Eating a handful of nuts daily can improve heart health.",
+    "Reducing salt intake can help lower blood pressure.",
+    "Aim for 7-8 hours of quality sleep each night for heart health.",
+    "Quitting smoking can cut your heart disease risk in half within a year.",
+    "Managing stress through meditation can benefit your cardiovascular system."
+]
+
+# Load models
+def load_ml_model(model_path):
+    return joblib.load(model_path)
+
+def load_dl_model(model_path):
+    return load_model(model_path)
+
+# Load scaler
+scaler = load_ml_model("models/scaler.sav")
+
+# Page config
+st.set_page_config(page_title="CardioPredictor", page_icon="❤️")
+
+def main():
+    st.title("Cardiovascular Risk Predictor App ❤️")
+    
+    # Display random health tip
+    st.info(f"💡 Health Tip: {random.choice(HEALTH_TIPS)}")
+    
+    # Create input form
+    with st.form("prediction_form"):
+        st.subheader("Patient Information")
+        
+        col1, col2 = st.columns(2)
+        
+        with col1:
+            age = st.number_input("Age", min_value=20, max_value=100, value=50)
+            sex = st.selectbox("Sex", ["Male", "Female"])
+            cp = st.selectbox("Chest Pain Type", [
+                "Typical angina", 
+                "Atypical angina", 
+                "Non-anginal pain", 
+                "Asymptomatic"
+            ])
+            trestbps = st.number_input("Resting Blood Pressure (mm Hg)", min_value=80, max_value=200, value=120)
+            chol = st.number_input("Serum Cholesterol (mg/dl)", min_value=100, max_value=600, value=200)
+            fbs = st.selectbox("Fasting Blood Sugar > 120 mg/dl", ["No", "Yes"])
+        
+        with col2:
+            restecg = st.selectbox("Resting ECG Results", [
+                "Normal",
+                "ST-T wave abnormality",
+                "Left ventricular hypertrophy"
+            ])
+            thalach = st.number_input("Maximum Heart Rate Achieved", min_value=60, max_value=220, value=150)
+            exang = st.selectbox("Exercise Induced Angina", ["No", "Yes"])
+            oldpeak = st.number_input("ST Depression Induced by Exercise", min_value=0.0, max_value=6.2, value=1.0, step=0.1)
+            slope = st.selectbox("Slope of Peak Exercise ST Segment", [
+                "Upsloping",
+                "Flat",
+                "Downsloping"
+            ])
+            ca = st.selectbox("Number of Major Vessels Colored by Fluoroscopy", [0, 1, 2, 3])
+            thal = st.selectbox("Thalassemia", [
+                "Normal",
+                "Fixed defect",
+                "Reversible defect"
+            ])
+        
+        model_choice = st.selectbox("Select Prediction Model", [
+            "Logistic Regression",
+            "Artificial Neural Network (ANN)",
+            "Long Short-Term Memory (LSTM)",
+            "XGBoost"
+        ])
+        
+        submitted = st.form_submit_button("Predict")
+    
+    if submitted:
+        # Convert inputs to model format
+        sex = 1 if sex == "Male" else 0
+        cp_mapping = {
+            "Typical angina": 0,
+            "Atypical angina": 1,
+            "Non-anginal pain": 2,
+            "Asymptomatic": 3
+        }
+        cp = cp_mapping[cp]
+        
+        fbs = 1 if fbs == "Yes" else 0
+        
+        restecg_mapping = {
+            "Normal": 0,
+            "ST-T wave abnormality": 1,
+            "Left ventricular hypertrophy": 2
+        }
+        restecg = restecg_mapping[restecg]
+        
+        exang = 1 if exang == "Yes" else 0
+        
+        slope_mapping = {
+            "Upsloping": 0,
+            "Flat": 1,
+            "Downsloping": 2
+        }
+        slope = slope_mapping[slope]
+        
+        thal_mapping = {
+            "Normal": 1,
+            "Fixed defect": 2,
+            "Reversible defect": 3
+        }
+        thal = thal_mapping[thal]
+        
+        # Prepare input data
+        input_data = {
+            'age': age,
+            'sex': sex,
+            'cp': cp,
+            'trestbps': trestbps,
+            'chol': chol,
+            'fbs': fbs,
+            'restecg': restecg,
+            'thalach': thalach,
+            'exang': exang,
+            'oldpeak': oldpeak,
+            'slope': slope,
+            'ca': ca,
+            'thal': thal
+        }
+        
+        # Preprocess input
+        processed_input = scaler.transform(pd.DataFrame([input_data]))
+        
+        try:
+            # Load model and make prediction
+            if model_choice == "Logistic Regression":
+                model = load_ml_model("models/logistic_model.sav")
+                proba = model.predict_proba(processed_input)[0]
+                prediction = model.predict(processed_input)
+                
+            elif model_choice == "Artificial Neural Network (ANN)":
+                model = load_dl_model("models/ann_model.h5")
+                proba = model.predict(processed_input)[0][0]
+                proba = [1-proba, proba]  # Convert to [prob_0, prob_1]
+                prediction = [1 if proba[1] > 0.5 else 0]
+                
+            elif model_choice == "Long Short-Term Memory (LSTM)":
+                model = load_dl_model("models/lstm_model.h5")
+                lstm_input = processed_input.reshape(1, 1, processed_input.shape[1])
+                proba = model.predict(lstm_input)[0][0]
+                proba = [1-proba, proba]  # Convert to [prob_0, prob_1]
+                prediction = [1 if proba[1] > 0.5 else 0]
+                
+            elif model_choice == "XGBoost":
+                model = load_ml_model("models/xgb_model.sav")
+                proba = model.predict_proba(processed_input)[0]
+                prediction = model.predict(processed_input)
+
+            # 1. Probability Output
+            st.subheader("Prediction Results")
+            if prediction[0] == 1:
+                st.error(f"High Risk of Cardiovascular Disease (Probability: {proba[1]:.2%})")
+            else:
+                st.success(f"Low Risk of Cardiovascular Disease (Probability: {proba[0]:.2%})")
+
+            # 2. Evaluation Metrics
+            st.subheader("Model Performance Metrics")
+            test_data = pd.read_csv("heart.csv")
+            X_test = test_data.drop('target', axis=1)
+            y_test = test_data['target']
+            X_test_scaled = scaler.transform(X_test)
+            
+            if model_choice in ["Logistic Regression", "XGBoost"]:
+                y_pred = model.predict(X_test_scaled)
+            elif model_choice == "Artificial Neural Network (ANN)":
+                y_pred = (model.predict(X_test_scaled) > 0.5).astype(int).flatten()
+            elif model_choice == "Long Short-Term Memory (LSTM)":
+                lstm_test = X_test_scaled.reshape(X_test_scaled.shape[0], 1, X_test_scaled.shape[1])
+                y_pred = (model.predict(lstm_test) > 0.5).astype(int).flatten()
+            
+            # Calculate metrics
+            st.write(f"Accuracy: {accuracy_score(y_test, y_pred):.2%}")
+            st.write(f"Precision: {precision_score(y_test, y_pred):.2%}")
+            st.write(f"Recall: {recall_score(y_test, y_pred):.2%}")
+            st.write(f"F1 Score: {f1_score(y_test, y_pred):.2%}")
+
+        except Exception as e:
+            st.error(f"Error during prediction: {str(e)}")
+
+if __name__ == '__main__':
     main()