Update app.py

app.py

@@ -1,25 +1,31 @@
-from fastapi import FastAPI, File, UploadFile, HTTPException
+from fastapi import FastAPI, File, UploadFile
+from fastapi.responses import JSONResponse
 from fastapi.middleware.cors import CORSMiddleware
-from tensorflow.keras.models import load_model
+import uvicorn
 import numpy as np
+import tensorflow as tf
+from tensorflow.keras.models import load_model
+from tensorflow.keras.preprocessing import image
 from PIL import Image
-import io
-import os
+from io import BytesIO
 
+# === Inisialisasi FastAPI ===
 app = FastAPI()
 
-# Untuk mengizinkan akses dari frontend kamu (misal domain lain)
+# === Middleware CORS agar bisa diakses dari frontend (JS) ===
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=["*"],  # Ganti dengan domain frontend-mu kalau sudah pasti
+    allow_origins=["*"],  # Ganti dengan domain frontend-mu untuk keamanan
     allow_credentials=True,
     allow_methods=["*"],
     allow_headers=["*"],
 )
 
-MODEL_PATH = 'model_cnn.h5'
+# === Konfigurasi Model ===
+MODEL_PATH = 'model_cnn.h5'  # Pastikan path model benar
 IMG_HEIGHT = 224
 IMG_WIDTH = 224
+
 class_names = [
     'freshapples', 'freshbanana', 'freshbittergroud', 'freshcapsicum', 'freshcucumber',
     'freshokra', 'freshoranges', 'freshpotato', 'freshtomato',
@@ -27,33 +33,47 @@ class_names = [
     'rottenokra', 'rottenoranges', 'rottenpotato', 'rottentomato'
 ]
 
-try:
-    full_model_path = os.path.join(os.getcwd(), MODEL_PATH)
-    model = load_model(full_model_path)
-    print(f"Model loaded successfully from: {full_model_path}")
-except Exception as e:
-    print(f"Error loading model: {e}")
-    raise RuntimeError(f"Failed to load model: {e}")
+# Muat model sekali saja saat startup
+model = load_model(MODEL_PATH)
+
+# === Fungsi bantu ===
+def read_imagefile(file) -> Image.Image:
+    image = Image.open(BytesIO(file)).convert("RGB")  # Pastikan RGB
+    return image
+
+def predict(img: Image.Image):
+    # Resize dan ubah ke array tanpa normalisasi ulang (karena model sudah punya Rescaling)
+    img = img.resize((IMG_WIDTH, IMG_HEIGHT))
+    img_array = image.img_to_array(img)  # [0,255]
+    img_array = np.expand_dims(img_array, axis=0)  # [1, 224, 224, 3]
+
+    prediction = model.predict(img_array)
+    predicted_class = np.argmax(prediction[0])
+    confidence = float(prediction[0][predicted_class]) * 100
 
+    return class_names[predicted_class], confidence
+
+# === Endpoint utama ===
 @app.post("/predict")
 async def predict_image(file: UploadFile = File(...)):
     try:
-        contents = await file.read()
-        image_pil = Image.open(io.BytesIO(contents)).convert("RGB")
-        image_pil = image_pil.resize((IMG_WIDTH, IMG_HEIGHT))
-        img_array = np.array(image_pil)
-        img_array = np.expand_dims(img_array, axis=0)
-
-        prediction = model.predict(img_array)
-        predicted_class_index = np.argmax(prediction[0])
-        confidence = float(prediction[0][predicted_class_index]) * 100
-        pred_class_name = class_names[predicted_class_index]
-
-        return {
-            "title": pred_class_name,
+        img = read_imagefile(await file.read())
+        pred_class, confidence = predict(img)
+
+        response = {
+            "filename": file.filename,
+            "title": "Prediction Result",
+            "message": f"The item is classified as: {pred_class}",
             "confidence": f"{confidence:.2f}",
-            "message": f"Prediksi: {pred_class_name} dengan kepercayaan {confidence:.2f}%",
-            "details": []
+            "details": [f"Class: {pred_class}", f"Confidence: {confidence:.2f}%"]
         }
+
+        return JSONResponse(content=response)
     except Exception as e:
-        raise HTTPException(status_code=500, detail=f"Error processing image: {str(e)}")
+        return JSONResponse(content={"error": str(e)}, status_code=500)
+
+# === Untuk menjalankan secara lokal ===
+if __name__ == "__main__":
+    uvicorn.run("app:app", host="0.0.0.0", port=8000, reload=True)
+
+# uvicorn app:app --reload