Update app.py

app.py

@@ -7,21 +7,20 @@ from typing import Tuple, List, Dict
 
 def load_model() -> Tuple[torch.nn.Module, List[str]]:
     """
-    Загружает модель и метки классов.
+    Loads the model and class labels.
 
     Returns:
-        model: Загруженная PyTorch модель.
-        labels: Список меток классов.
+        model: The loaded PyTorch model.
+        labels: List of class labels.
     """
     model_path = "skin_disease_model_jit.pt"
     labels_path = "labels.txt"
     
     if not os.path.exists(model_path):
-        raise FileNotFoundError(f"Модель не найдена: {model_path}")
+        raise FileNotFoundError(f"Model not found: {model_path}")
     if not os.path.exists(labels_path):
-        raise FileNotFoundError("Файл labels.txt не найден.")
+        raise FileNotFoundError("File labels.txt not found.")
     
-    # Если доступна GPU, используем её, иначе CPU.
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     model = torch.jit.load(model_path, map_location=device)
     model.eval()
@@ -33,7 +32,7 @@ def load_model() -> Tuple[torch.nn.Module, List[str]]:
 
 model, labels = load_model()
 
-# Определение преобразований для входного изображения.
+# Define image preprocessing steps
 preprocess = transforms.Compose([
     transforms.Resize((224, 224)),
     transforms.ToTensor(),
@@ -43,27 +42,28 @@ preprocess = transforms.Compose([
 
 def predict(image: Image.Image) -> Dict[str, float]:
     """
-    Выполняет предсказание для переданного изображения.
+    Makes a prediction for the given image.
 
     Args:
-        image (PIL.Image): Изображение для анализа.
+        image (PIL.Image): The input image.
     
     Returns:
-        Dict[str, float]: Словарь, где ключ — класс, значение — вероятность.
+        Dict[str, float]: A dictionary where keys are class names, and values are probabilities.
     """
     try:
         image = image.convert("RGB")
         image_tensor = preprocess(image).unsqueeze(0)
-        # Определяем устройство для обработки.
         device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
         image_tensor = image_tensor.to(device)
         model.to(device)
+        
         with torch.no_grad():
             output = model(image_tensor)
             scores = torch.nn.functional.softmax(output[0], dim=0)
+        
         predictions = {label: float(score) for label, score in zip(labels, scores)}
-        # Сортировка предсказаний по убыванию вероятности.
         sorted_predictions = dict(sorted(predictions.items(), key=lambda item: item[1], reverse=True))
+        
         return sorted_predictions
     except Exception as e:
         return {"error": str(e)}
@@ -73,7 +73,7 @@ description = (
     "🔬 **Skin-AI — AI-Powered Skin Disease Classification**\n\n"
     "This project utilizes a deep learning model to classify skin diseases based on an uploaded image.\n\n"
     "### 🚀 How to Use:\n\n"
-    "1️⃣ Upload or take a photo of the affected skin area. \n\n"
+    "1️⃣ Upload or take a photo of the affected skin area.\n\n"
     "2️⃣ Click the 'Submit' button.\n\n"
     "3️⃣ The app will return the probabilities for possible skin conditions.\n\n"
     "⚠️ **Important!** The results are for informational purposes only and do not constitute a medical diagnosis.\n\n"
@@ -81,15 +81,22 @@ description = (
     "- PyTorch (Lightning)\n"
     "- Gradio\n"
     "- Hugging Face Spaces\n\n"
-    "🔗 Source Code: [GitHub/Hugging Face](https://huggingface.co/spaces/Eraly-ml/Skin-AI)"
+    "🔗 Source Code: [GitHub/Hugging Face](https://huggingface.co/spaces/Eraly-ml/Skin-A)"
 )
 
+# Adding example images
+examples = [
+    ["example1.jpg"],
+    ["example2.jpg"]
+]
+
 interface = gr.Interface(
     fn=predict,
-    inputs=gr.Image(type="pil", label="Image"),
+    inputs=gr.Image(type="pil", label="Upload Image"),
     outputs=gr.Label(num_top_classes=3, label="Prediction"),
     title=title,
     description=description,
+    examples=examples,
     theme=gr.themes.Soft()
 )