Update app.py

app.py

@@ -1,90 +1,99 @@
 import streamlit as st
 import os
-import subprocess
-
-# --- 1. JAVA VE ÇEVRESEL AYARLAR (JAVA & ENVIRONMENT SETTINGS) ---
-try:
-    # Java yolunu sistemden otomatik bulur (Finds Java path automatically)
-    java_path = subprocess.check_output(['which', 'java']).decode('utf-8').strip()
-    real_java_path = os.path.realpath(java_path)
-    os.environ["JAVA_HOME"] = real_java_path.replace("/bin/java", "")
-except:
-    # Hata durumunda varsayılan yol (Default path if fails)
-    os.environ["JAVA_HOME"] = "/usr/lib/jvm/java-17-openjdk-amd64"
-
 from pyspark.sql import SparkSession
 from pyspark.ml.classification import LogisticRegressionModel
 from pyspark.ml.feature import VectorAssembler
 
-# --- 2. SAYFA AYARLARI (PAGE SETTINGS) ---
-st.set_page_config(page_title="Heart Disease Predictor", page_icon="🫀")
+# --- 1. SAYFA AYARLARI (PAGE CONFIG) ---
+st.set_page_config(page_title="Heart Disease Predictor", page_icon="🫀", layout="centered")
+
+# CSS ile buton ve başlıkları güzelleştirelim
+st.markdown("""
+    <style>
+    .main { opacity: 0.95; }
+    .stButton>button { width: 100%; border-radius: 20px; height: 3em; background-color: #ff4b4b; color: white; }
+    </style>
+    """, unsafe_allow_html=True)
+
 st.title("🫀 Heart Disease Risk Prediction")
 st.subheader("Kalp Hastalığı Risk Tahmini")
+st.write("---")
 
-# --- 3. SPARK OTURUMU (SPARK SESSION) ---
+# --- 2. SPARK VE MODEL YÜKLEME (SPARK & MODEL LOADING) ---
 @st.cache_resource
-def get_spark():
-    return SparkSession.builder \
-        .appName("HeartDiseasePredictor") \
+def initialize_app():
+    # Spark oturumunu başlat
+    spark = SparkSession.builder \
+        .appName("HeartDiseaseApp") \
         .master("local[*]") \
+        .config("spark.driver.bindAddress", "127.0.0.1") \
         .getOrCreate()
-
-spark = get_spark()
-
-# --- 4. MODELİ YÜKLEME (LOAD MODEL) ---
-@st.cache_resource
-def load_heart_model():
-    # heart_model klasöründen verileri çeker
-    return LogisticRegressionModel.load("heart_model")
+    
+    # Modeli yükle (heart_model klasöründen)
+    model = LogisticRegressionModel.load("heart_model")
+    return spark, model
 
 try:
-    model = load_heart_model()
-    st.success("✅ Model loaded successfully! / Model başarıyla yüklendi!")
+    spark, model = initialize_app()
+    st.success("✅ System Ready / Sistem Hazır")
 except Exception as e:
-    st.error(f"❌ Error loading model: {e}")
+    st.error(f"❌ Initialization Error / Başlatma Hatası: {e}")
+    st.info("Lütfen heart_model klasörünün doğruluğunu kontrol edin.")
     st.stop()
 
-# --- 5. GİRİŞ ALANLARI (INPUT FIELDS) ---
-st.sidebar.header("Patient Data / Hasta Verileri")
+# --- 3. KULLANICI GİRİŞLERİ (USER INPUTS) ---
+st.sidebar.header("📋 Patient Data / Hasta Verileri")
 
-def user_input_features():
-    age = st.sidebar.number_input("Age / Yaş", 1, 120, 50)
-    sex = st.sidebar.selectbox("Sex / Cinsiyet (1: M, 0: F)", [1, 0])
-    cp = st.sidebar.selectbox("Chest Pain / Göğüs Ağrısı (0-3)", [0, 1, 2, 3])
-    trestbps = st.sidebar.number_input("Resting Blood Pressure / Kan Basıncı", 50, 250, 120)
-    chol = st.sidebar.number_input("Cholesterol / Kolesterol", 100, 600, 200)
-    fbs = st.sidebar.selectbox("Fasting Blood Sugar > 120 (1: Y, 0: N)", [0, 1])
-    restecg = st.sidebar.selectbox("Resting ECG / Dinlenme EKG (0-2)", [0, 1, 2])
-    thalach = st.sidebar.number_input("Max Heart Rate / Maks. Kalp Hızı", 50, 250, 150)
-    exang = st.sidebar.selectbox("Exercise Induced Angina (1: Y, 0: N)", [0, 1])
-    oldpeak = st.sidebar.number_input("Oldpeak", 0.0, 10.0, 1.0)
-    slope = st.sidebar.selectbox("Slope (0-2)", [0, 1, 2])
-    ca = st.sidebar.selectbox("Major Vessels / Damar Sayısı (0-4)", [0, 1, 2, 3, 4])
-    thal = st.sidebar.selectbox("Thal (0-3)", [0, 1, 2, 3])
+def get_inputs():
+    col1, col2 = st.columns(2)
     
-    data = {
+    with col1:
+        age = st.number_input("Age / Yaş", 1, 120, 50)
+        sex = st.selectbox("Sex / Cinsiyet", options=[1, 0], format_func=lambda x: "Male/Erkek (1)" if x == 1 else "Female/Kadın (0)")
+        cp = st.selectbox("Chest Pain / Göğüs Ağrısı (0-3)", [0, 1, 2, 3])
+        trestbps = st.number_input("Resting BP / Kan Basıncı", 50, 250, 120)
+        chol = st.number_input("Cholesterol / Kolesterol", 100, 600, 200)
+        fbs = st.selectbox("Fasting Sugar > 120 / Şeker", [0, 1], format_func=lambda x: "Yes/Evet (1)" if x == 1 else "No/Hayır (0)")
+        
+    with col2:
+        restecg = st.selectbox("Resting ECG / EKG (0-2)", [0, 1, 2])
+        thalach = st.number_input("Max Heart Rate / Kalp Hızı", 50, 250, 150)
+        exang = st.selectbox("Exercise Angina / Anjin", [0, 1], format_func=lambda x: "Yes/Evet (1)" if x == 1 else "No/Hayır (0)")
+        oldpeak = st.number_input("Oldpeak", 0.0, 10.0, 1.0)
+        slope = st.selectbox("Slope (0-2)", [0, 1, 2])
+        ca = st.selectbox("Major Vessels / Damar Sayısı", [0, 1, 2, 3, 4])
+        thal = st.selectbox("Thal (0-3)", [0, 1, 2, 3])
+
+    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
     }
-    return spark.createDataFrame([data])
+    return spark.createDataFrame([input_data])
 
-input_df = user_input_features()
+input_df = get_inputs()
 
-# --- 6. TAHMİN (PREDICTION) ---
-if st.button("Predict / Tahmin Et"):
-    feature_cols = ['age', 'sex', 'cp', 'trestbps', 'chol', 'fbs', 'restecg', 'thalach', 'exang', 'oldpeak', 'slope', 'ca', 'thal']
-    assembler = VectorAssembler(inputCols=feature_cols, outputCol="features")
-    final_data = assembler.transform(input_df)
-    
-    prediction = model.transform(final_data)
-    result = prediction.select("prediction").collect()[0][0]
-    probability = prediction.select("probability").collect()[0][0]
+# --- 4. TAHMİN (PREDICTION) ---
+st.write("---")
+if st.button("PREDICT / TAHMİN ET"):
+    with st.spinner('Analyzing... / Analiz ediliyor...'):
+        # Özellikleri vektöre dönüştür
+        feature_cols = ['age', 'sex', 'cp', 'trestbps', 'chol', 'fbs', 'restecg', 'thalach', 'exang', 'oldpeak', 'slope', 'ca', 'thal']
+        assembler = VectorAssembler(inputCols=feature_cols, outputCol="features")
+        final_data = assembler.transform(input_df)
+        
+        # Tahmin yap
+        prediction_results = model.transform(final_data)
+        result = prediction_results.select("prediction").collect()[0][0]
+        probability = prediction_results.select("probability").collect()[0][0]
 
-    if result == 1.0:
-        st.error(f"⚠️ Risk: HIGH / YÜKSEK (Prob: %{round(probability[1]*100, 2)})")
-    else:
-        st.success(f"💚 Risk: LOW / DÜŞÜK (Prob: %{round(probability[0]*100, 2)})")
+        st.subheader("Results / Sonuçlar")
+        if result == 1.0:
+            st.error(f"⚠️ HIGH RISK / YÜKSEK RİSK")
+            st.write(f"Confidence / Güven Oranı: %{round(probability[1]*100, 2)}")
+        else:
+            st.success(f"💚 LOW RISK / DÜŞÜK RİSK")
+            st.write(f"Confidence / Güven Oranı: %{round(probability[0]*100, 2)}")
 
-st.divider()
-st.caption("Disclaimer: This is for educational purposes only. / Tıbbi tavsiye değildir.")
+st.write("---")
+st.caption("Disclaimer: Not for medical use. / Tıbbi amaçlı kullanılamaz.")