Update app.py

app.py

@@ -1,12 +1,12 @@
 import gradio as gr
 import torch
+import random
+import requests
+import os
 from torchvision import transforms
 from PIL import Image
 
-# اسم الريبو الخاص بك على Hugging Face
 REPO_ID = "Alhdrawi/x_alhdrawi"
-
-# أسماء الملفات المرفوعة
 MODEL_FILES = [
     "best_128_0.0002_original_15000_0.859.pt",
     "best_128_0.0002_original_8000_0.857.pt",
@@ -20,34 +20,44 @@ MODEL_FILES = [
     "best_64_5e-05_original_22000_0.864.pt",
 ]
 
-# نفس المعالجة المستخدمة
-transform = transforms.Compose([
-    transforms.Resize((224, 224)),
-    transforms.ToTensor(),
-    transforms.Normalize(mean=[0.485], std=[0.229])
-])
-
-# الأمراض اللي يشخصها
 diseases = [
     "Atelectasis", "Cardiomegaly", "Consolidation", "Edema", "Effusion",
     "Emphysema", "Fibrosis", "Hernia", "Infiltration", "Mass", "Nodule",
     "Pleural_Thickening", "Pneumonia", "Pneumothorax"
 ]
 
+transform = transforms.Compose([
+    transforms.Resize((224, 224)),
+    transforms.ToTensor(),
+    transforms.Normalize(mean=[0.485], std=[0.229])
+])
+
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 model = None
+current_model_path = None
+
+def download_model_file(filename):
+    url = f"https://huggingface.co/{REPO_ID}/resolve/main/{filename}"
+    local_path = f"./{filename}"
+    if not os.path.exists(local_path):
+        print(f"Downloading model from {url}")
+        response = requests.get(url)
+        with open(local_path, "wb") as f:
+            f.write(response.content)
+    return local_path
 
-def load_model(checkpoint_file):
-    global model
-    print(f"Loading model: {checkpoint_file}")
-    model_url = f"https://huggingface.co/{REPO_ID}/resolve/main/{checkpoint_file}"
-    model = torch.load(model_url, map_location=device)
+def load_model(_):
+    global model, current_model_path
+    selected_model = random.choice(MODEL_FILES)
+    local_path = download_model_file(selected_model)
+    model = torch.load(local_path, map_location=device)
     model.eval()
-    return f"تم تحميل النموذج: {checkpoint_file}"
+    current_model_path = selected_model
+    return f"تم تحميل النموذج العشوائي: {selected_model}"
 
 def predict(image):
     if model is None:
-        return "الرجاء تحميل نموذج أولاً"
+        return "❗ الرجاء تحميل نموذج أولاً"
     img = transform(image.convert("L")).unsqueeze(0).to(device)
     with torch.no_grad():
         outputs = model(img)
@@ -56,16 +66,15 @@ def predict(image):
     return results
 
 with gr.Blocks() as demo:
-    gr.Markdown("## 🧠 CheXzero | اختر النموذج وارفع صورة أشعة")
+    gr.Markdown("## 🧠 CheXzero | اختر صورة أشعة لتحليلها عبر نموذج يتم تحميله عشوائيًا")
     with gr.Row():
-        model_selector = gr.Dropdown(label="اختر نموذج", choices=MODEL_FILES, value=MODEL_FILES[-1])
-        load_button = gr.Button("تحميل النموذج")
+        load_button = gr.Button("تحميل نموذج عشوائي")
         load_status = gr.Textbox(label="الحالة", interactive=False)
     with gr.Row():
         image_input = gr.Image(type="pil", label="صورة الأشعة")
         output = gr.Label(num_top_classes=5)
 
-    load_button.click(fn=load_model, inputs=model_selector, outputs=load_status)
+    load_button.click(fn=load_model, inputs=None, outputs=load_status)
     image_input.change(fn=predict, inputs=image_input, outputs=output)
 
 demo.launch()