Update app.py

app.py

@@ -1,89 +1,44 @@
 import streamlit as st
-import torch
+from ultralytics import YOLO
 from PIL import Image
 import numpy as np
-import cv2
-import matplotlib.pyplot as plt
 
-# Konfigurasi model
-MODEL_PATH = "best.pt"  # Pastikan model ada di direktori ini
-CLASS_NAMES = ["bag", "person-static-object"]  # Sesuaikan dengan kelas Anda
+# Konfigurasi Model
+MODEL_PATH = "best.pt"  # Pastikan model sudah diupload ke repo
+CLASS_NAMES = ["bag", "person-static-object"]
 
-# Fungsi untuk memuat model
+# Muat Model
 @st.cache_resource
 def load_model():
-    model = torch.load(MODEL_PATH, map_location=torch.device('cpu'))
-    model.eval()  # Mode evaluasi
+    model = YOLO(MODEL_PATH)  # Load model YOLOv11
     return model
 
-# Preprocessing gambar
-def preprocess(image):
-    # Resize dan konversi ke tensor
-    image = np.array(image)
-    image = cv2.resize(image, (640, 640))  # Sesuaikan dengan input size model
-    image = torch.from_numpy(image).permute(2, 0, 1).float() / 255.0
-    image = image.unsqueeze(0)  # Tambahkan batch dimension
-    return image
+# Streamlit UI
+st.title("YOLOv11 Object Detector 🚀")
+st.write("Deteksi objek 'bag' dan 'person-static-object' menggunakan YOLOv11")
 
-# Postprocessing hasil prediksi
-def postprocess(prediction, confidence_threshold=0.5):
-    # Contoh untuk deteksi objek YOLO-style (sesuaikan dengan output model Anda)
-    boxes = prediction[0]["boxes"].detach().numpy()
-    scores = prediction[0]["scores"].detach().numpy()
-    labels = prediction[0]["labels"].detach().numpy()
-    
-    # Filter berdasarkan confidence threshold
-    keep = scores >= confidence_threshold
-    return boxes[keep], scores[keep], labels[keep]
-
-# Antarmuka Streamlit
-st.title("Deteksi Objek Tertinggal 👜👤")
-st.write("Upload gambar untuk mendeteksi objek 'bag' atau 'person-static-object'")
-
-# Upload gambar
-uploaded_file = st.file_uploader("Pilih gambar...", type=["jpg", "png", "jpeg"])
+uploaded_file = st.file_uploader("Upload gambar...", type=["jpg", "png", "jpeg"])
 
-if uploaded_file is not None:
-    # Memuat gambar
+if uploaded_file:
+    # Tampilkan gambar input
     image = Image.open(uploaded_file).convert("RGB")
     st.image(image, caption="Gambar Input", use_container_width=True)
     
-    # Proses deteksi
     if st.button("Deteksi Objek"):
         with st.spinner("Memproses..."):
             try:
-                # Memuat model
+                # Proses deteksi
                 model = load_model()
-                
-                # Preprocessing
-                input_tensor = preprocess(image)
-                
-                # Prediksi
-                with torch.no_grad():
-                    prediction = model(input_tensor)
-                
-                # Postprocessing
-                boxes, scores, labels = postprocess(prediction)
+                results = model.predict(image, conf=0.5)
                 
                 # Visualisasi hasil
-                fig, ax = plt.subplots(1, figsize=(12, 6))
-                ax.imshow(image)
+                res_plotted = results[0].plot()[:, :, ::-1]  # Konversi BGR ke RGB
+                st.image(res_plotted, caption="Hasil Deteksi", use_container_width=True)
                 
-                for box, score, label in zip(boxes, scores, labels):
-                    x1, y1, x2, y2 = box
-                    rect = plt.Rectangle(
-                        (x1, y1), x2 - x1, y2 - y1,
-                        linewidth=2, edgecolor="lime", facecolor="none"
-                    )
-                    ax.add_patch(rect)
-                    ax.text(
-                        x1, y1 - 5,
-                        f"{CLASS_NAMES[label]}: {score:.2f}",
-                        color="lime", fontsize=10, backgroundcolor="black"
-                    )
+                # Tampilkan metrik
+                boxes = results[0].boxes
+                st.success(f"✅ Objek Terdeteksi: {len(boxes)}")
+                st.write(f"**mAP@0.5:** {results[0].speed['mAP50']:.3f}")
                 
-                st.pyplot(fig)
-                st.success("Selesai!")
-            
             except Exception as e:
-                st.error(f"Error: {str(e)}")
+                st.error(f"❌ Error: {str(e)}")