Update app.py

app.py

@@ -1,36 +1,32 @@
 import gradio as gr
 import tensorflow as tf
-from tensorflow.keras.applications.resnet import ResNet152, preprocess_input, decode_predictions
+from tensorflow.keras.applications import EfficientNetV2L
+from tensorflow.keras.applications.efficientnet_v2 import preprocess_input, decode_predictions
 from tensorflow.keras.preprocessing.image import img_to_array
 from PIL import Image
 import numpy as np
 
-# Load the pre-trained ResNet152 model
-MODEL_PATH = "resnet152-image-classifier.h5"  # Path to the saved model
-try:
-    model = tf.keras.models.load_model(MODEL_PATH)
-except Exception as e:
-    print(f"Error loading model: {e}")
-    exit()
+# Load a stronger pretrained model (EfficientNetV2L)
+model = EfficientNetV2L(weights="imagenet")
 
 def predict_image(image):
     """
-    Process the uploaded image and return the top 3 predictions.
+    Process the uploaded image and return the top 5 predictions.
     """
     try:
         # Preprocess the image
-        image = image.resize((224, 224))  # ResNet152 expects 224x224 input
+        image = image.resize((480, 480))  # EfficientNetV2L expects 480x480 input
         image_array = img_to_array(image)
         image_array = preprocess_input(image_array)  # Normalize the image
         image_array = np.expand_dims(image_array, axis=0)  # Add batch dimension
 
         # Get predictions
         predictions = model.predict(image_array)
-        decoded_predictions = decode_predictions(predictions, top=3)[0]
+        decoded_predictions = decode_predictions(predictions, top=5)[0]
 
-        # Format predictions as a list of tuples (label, confidence)
-        results = [(label, float(confidence)) for _, label, confidence in decoded_predictions]
-        return dict(results)
+        # Format predictions as a dictionary (label -> confidence)
+        results = {label: float(confidence) for _, label, confidence in decoded_predictions}
+        return results
 
     except Exception as e:
         return {"Error": str(e)}
@@ -39,12 +35,12 @@ def predict_image(image):
 interface = gr.Interface(
     fn=predict_image,
     inputs=gr.Image(type="pil"),  # Accepts an image input
-    outputs=gr.Label(num_top_classes=3),  # Shows top 3 predictions with confidence
-    title="ResNet152 Image Classifier",
-    description="Upload an image, and the model will predict what's in the image.",
-    examples=["dog.jpg", "cat.jpg"],  # Example images for users to test
+    outputs=gr.Label(num_top_classes=2),  # Shows top 5 predictions with confidence
+    title="Image Classifier",
+    description="Upload an image, and the model will predict what's in the image with higher accuracy.",
+    examples=["dog.jpg", "cat.jpg", "building.jpg", "tree.jpg"],  # Example images for testing
 )
 
 # Launch the Gradio app
 if __name__ == "__main__":
-    interface.launch()
+    interface.launch()