Update app.py

app.py

@@ -6,10 +6,10 @@ from PIL import Image
 from torchvision import transforms
 import scipy.io
 from pytorch_grad_cam import GradCAMPlusPlus
-from pytorch_grad_cam.utils.image import show_cam_on_image, preprocess_image
+from pytorch_grad_cam.utils.image import show_cam_on_image
 from pytorch_grad_cam.utils.model_targets import ClassifierOutputTarget
 
-# Load model and class names
+# --- Load model and metadata ---
 MODEL_PATH = "efficientnetv2_best_model.pth"
 META_PATH = "cars_meta.mat"
 DEVICE = torch.device("cpu")
@@ -36,7 +36,7 @@ model.load_state_dict(state_dict)
 model.eval()
 model.to(DEVICE)
 
-# Grad-CAM++
+# --- Grad-CAM++ prediction function ---
 def predict_and_explain(img):
     image_pil = img.convert("RGB").resize((224, 224))
     input_tensor = val_transform(image_pil).unsqueeze(0).to(DEVICE)
@@ -46,6 +46,7 @@ def predict_and_explain(img):
         pred_idx = output.argmax(dim=1).item()
         pred_name = class_names[pred_idx]
 
+    # Grad-CAM++
     targets = [ClassifierOutputTarget(pred_idx)]
     cam = GradCAMPlusPlus(model=model, target_layers=[model.blocks[-1][-1]])
     grayscale_cam = cam(input_tensor=input_tensor, targets=targets)[0]
@@ -56,11 +57,14 @@ def predict_and_explain(img):
     cam_pil = Image.fromarray(cam_image)
     return cam_pil, f"Prediction: {pred_name} (class index {pred_idx})"
 
-# Gradio Interface
+# --- Gradio UI ---
 demo = gr.Interface(
     fn=predict_and_explain,
-    inputs=gr.Image(type="pil", label="Upload Car Image"),
-    outputs=[gr.Image(label="Grad-CAM++ Heatmap"), gr.Text(label="Prediction")],
+    inputs=gr.Image(type="pil", label="Upload Car Image", height=350),
+    outputs=[
+        gr.Image(type="pil", label="Grad-CAM++ Heatmap", height=350),
+        gr.Text(label="Prediction")
+    ],
     title="🚗 EfficientNetV2 Car Classifier + Grad-CAM++",
     description="Upload a car image to see its predicted make/model/year and what influenced the prediction.",
 )