Update app.py

app.py

@@ -37,22 +37,26 @@ model.eval()
 emotion_labels = ['Angry', 'Disgust', 'Fear', 'Happy', 'Sad', 'Surprise', 'Neutral']
 
 # Define preprocessing
+from PIL import ImageOps
+import torchvision.transforms as transforms
+
 transform = transforms.Compose([
-    transforms.Grayscale(),
+    transforms.Grayscale(num_output_channels=1),
     transforms.Resize((48, 48)),
     transforms.ToTensor(),
 ])
 
 def predict_emotion(image):
     if image is None:
-        return "No image uploaded."
-    try:
-        image = transform(image).unsqueeze(0).to(device)
-        output = model(image)
-        _, predicted = torch.max(output, 1)
-        return emotion_labels[predicted.item()]
-    except Exception as e:
-        return f"⚠️ Error: {e}"
+        return "No image provided"
+
+    image = ImageOps.grayscale(image)                # Convert to grayscale
+    image = image.resize((48, 48))                   # Resize to training shape
+    img_tensor = transform(image).unsqueeze(0).to(device)  # Transform and batch
+
+    output = model(img_tensor)
+    prediction = output.argmax(dim=1).item()
+    return EMOTIONS[prediction]
 
 # Launch Gradio interface
 demo = gr.Interface(