Creating an Endpoint

Dockerfile

@@ -13,4 +13,4 @@ COPY . .
 # Buka Port 7860 (Standar Hugging Face Spaces)
 EXPOSE 7860
 
-CMD ["streamlit", "run", "app.py", "--server.port", "7860", "--server.address", "0.0.0.0"]
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

app.py

@@ -1,62 +1,47 @@
-# ========================================
-# IMPORT LIBRARY
-# ========================================
-import streamlit as st
+from fastapi import FastAPI, File, UploadFile, HTTPException
 import tensorflow as tf
 import numpy as np
 from PIL import Image
-from tensorflow.keras.applications.resnet50 import preprocess_input
+import io
 import sys
+from tensorflow.keras.applications.resnet50 import preprocess_input
 
-# ========================================
-# 1. KONFIGURASI HALAMAN
-# ========================================
-st.set_page_config(
-    page_title="Ashoka Hipospadia Classifier",
-    page_icon="✨",
-    layout="centered"
-)
+# 1. Inisialisasi Aplikasi
+app = FastAPI(title="Ashoka Hipospadia Classifier API")
 
-# ========================================
-# 2. LOAD MODEL
-# ========================================
-@st.cache_resource
-def load_model():
-    """Load trained model dari file .h5"""
-    try:
-        print("Sedang memuat model...")
-        model = tf.keras.models.load_model('cnn_kfold_best_model.h5')
-        print("Model berhasil dimuat!")
-        return model
-    except Exception as e:
-        st.error(f"Error memuat model: {e}")
-        print(f"Error memuat model: {e}")
-        return None
-
-model = load_model()
+# 2. Load Model
+print("Sedang memuat model...")
+try:
+    model = tf.keras.models.load_model('cnn_kfold_best_model.h5')
+    print("Model berhasil dimuat!")
+except Exception as e:
+    print(f"Error memuat model: {e}")
+    sys.exit(1)  # Matikan server jika model gagal load
 
 # Label kelas: 0 = normal, 1 = buried
 class_names = ['normal', 'buried']
 
-# ========================================
-# 3. FUNGSI PREPROCESSING GAMBAR
-# ========================================
-def prepare_image(image):
+# 3. Fungsi Preprocessing
+def prepare_image(image_bytes):
     """
-    Preprocessing gambar sebelum prediksi
+    Preprocessing gambar untuk model ResNet50
     - Konversi ke RGB (3 channel)
     - Resize ke 224x224
     - Preprocessing ResNet50
     """
     try:
+        img = Image.open(io.BytesIO(image_bytes))
+        
         # Paksa ubah ke RGB agar PNG transparan tidak error
-        img = image.convert("RGB")
+        img = img.convert("RGB")
         
-        # Resize ke ukuran input model
+        # Resize ke ukuran input model (224x224 untuk ResNet50)
         img = img.resize((224, 224))
         
-        # Convert ke numpy array dan tambah batch dimension
+        # Convert ke numpy array
         img_array = np.array(img)
+        
+        # Tambah batch dimension
         img_array = np.expand_dims(img_array, axis=0)
         
         # Preprocessing ResNet50 (HARUS sama dengan training!)
@@ -67,111 +52,61 @@ def prepare_image(image):
         print(f"Error saat memproses gambar: {e}")
         return None
 
-# ========================================
-# 4. INTERFACE WEB
-# ========================================
-
-# Header
-st.title("✨ Ashoka Hipospadia Classifier")
-st.write("Upload an image for AI-powered classification")
-st.divider()
-
-# File uploader
-uploaded_file = st.file_uploader(
-    "📁 Choose an image",
-    type=['jpg', 'jpeg', 'png', 'bmp']
-)
-
-# ========================================
-# 5. LOGIKA PREDIKSI
-# ========================================
-if uploaded_file:
+# 4. Endpoint Prediksi
+@app.post("/predict")
+async def predict(file: UploadFile = File(...)):
+    """
+    Endpoint untuk prediksi gambar
+    Input: File gambar (JPG, PNG, BMP)
+    Output: JSON dengan class dan confidence
+    """
     try:
-        # Baca gambar
-        image = Image.open(uploaded_file)
+        # Baca file gambar
+        image_bytes = await file.read()
+        
+        # Proses gambar
+        processed_image = prepare_image(image_bytes)
+        
+        if processed_image is None:
+            raise HTTPException(status_code=400, detail="File bukan gambar yang valid")
+        
+        # Prediksi
+        prediction = model.predict(processed_image)
+        pred_value = float(prediction[0][0])
+        
+        # Hitung probabilitas
+        # Model output: 0 = normal, 1 = buried
+        prob_normal = (1 - pred_value) * 100
+        prob_buried = pred_value * 100
+        
+        # Tentukan kelas berdasarkan threshold 0.5
+        top_class_idx = 1 if pred_value > 0.5 else 0
         
-        # Tampilkan preview
-        st.image(image, caption="Uploaded Image", use_column_width=True)
+        # Hasil dalam format JSON
+        result = {
+            "class": class_names[top_class_idx],
+            "confidence": float(max(prob_normal, prob_buried)),
+            "probabilities": {
+                "normal": float(prob_normal),
+                "buried": float(prob_buried)
+            }
+        }
+        return result
         
-        # Tombol prediksi
-        if st.button("🔍 Analyze Image", type="primary", use_container_width=True):
-            with st.spinner("🔄 Processing image..."):
-                
-                # Preprocessing
-                processed_image = prepare_image(image)
-                
-                if processed_image is None:
-                    st.error("❌ File bukan gambar yang valid")
-                elif model is None:
-                    st.error("❌ Model gagal dimuat")
-                else:
-                    # Prediksi
-                    prediction = model.predict(processed_image, verbose=0)
-                    pred_value = float(prediction[0][0])
-                    
-                    # Hitung probabilitas
-                    prob_normal = (1 - pred_value) * 100
-                    prob_buried = pred_value * 100
-                    
-                    # Tentukan kelas (threshold 0.5)
-                    top_class_idx = 1 if pred_value > 0.5 else 0
-                    predicted_class = class_names[top_class_idx]
-                    confidence = max(prob_normal, prob_buried)
-                    
-                    # Hasil prediksi
-                    result = {
-                        "class": predicted_class,
-                        "confidence": confidence,
-                        "probabilities": {
-                            "normal": prob_normal,
-                            "buried": prob_buried
-                        }
-                    }
-                    
-                    # Tampilkan hasil
-                    st.divider()
-                    st.subheader("📊 Classification Results")
-                    
-                    # Status box
-                    if predicted_class == "normal":
-                        st.success(f"✅ Classification: **{predicted_class.upper()}**")
-                    else:
-                        st.error(f"⚠️ Classification: **{predicted_class.upper()}**")
-                    
-                    # Confidence metric
-                    st.metric(
-                        label="Confidence Level",
-                        value=f"{confidence:.1f}%",
-                        delta="High" if confidence > 80 else "Moderate"
-                    )
-                    
-                    # Detail probabilitas
-                    st.subheader("📈 Detailed Probability")
-                    col1, col2 = st.columns(2)
-                    
-                    with col1:
-                        st.metric("🟢 Normal", f"{prob_normal:.1f}%")
-                    
-                    with col2:
-                        st.metric("🔴 Buried", f"{prob_buried:.1f}%")
-                    
-                    # Visual progress bars
-                    st.write("**Visual Distribution:**")
-                    st.progress(prob_normal / 100, text=f"Normal: {prob_normal:.1f}%")
-                    st.progress(prob_buried / 100, text=f"Buried: {prob_buried:.1f}%")
-                    
     except Exception as e:
-        st.error(f"❌ CRITICAL ERROR: {e}")
+        # Cetak error ke log
         print(f"CRITICAL ERROR: {e}")
+        raise HTTPException(status_code=500, detail=str(e))
 
-else:
-    # ========================================
-    # EMPTY STATE (belum upload gambar)
-    # ========================================
-    st.info("👆 Please upload an image to begin analysis")
+# 5. Endpoint Home
+@app.get("/")
+def home():
+    """Endpoint root untuk testing API"""
+    return {
+        "message": "Ashoka Hipospadia Classifier API Online! 🚀",
+        "model": "ResNet50 Binary Classification",
+        "classes": class_names
+    }
 
-# ========================================
-# FOOTER
-# ========================================
-st.divider()
-st.caption("🚀 Powered by **Ashoka AI** | Deep Learning for Hipospadia Diagnosis")
+# API siap digunakan dengan uvicorn
+# Jalankan dengan: uvicorn app:app --host 0.0.0.0 --port 7860

requirements.txt

@@ -1,4 +1,6 @@
 tensorflow-cpu>=2.16.0
-streamlit
+fastapi
+uvicorn
+python-multipart
 pillow
-numpy
+numpy