Update app.py

app.py

@@ -1,20 +1,11 @@
 import streamlit as st
-import tensorflow as tf
+from transformers import TFAutoModelForImageClassification, AutoFeatureExtractor
 from PIL import Image
 import numpy as np
-import zipfile
-import os
 
-# Function to load the model from the zip file
-def load_model_from_zip(zip_file_path):
-    with zipfile.ZipFile(zip_file_path, 'r') as zip_ref:
-        zip_ref.extractall('.')  # Extract to the current directory
-    # Load the SavedModel directly from the current directory (root)
-    model = tf.keras.models.load_model('.')  
-    return model
-
-# Load the model 
-model = load_model_from_zip('my_christmas_tree_model.zip')  
+# Load the model and feature extractor
+model = TFAutoModelForImageClassification.from_pretrained("your-username/your-repo-name")  # Replace with your repo
+feature_extractor = AutoFeatureExtractor.from_pretrained(model.config._name_or_path)
 
 # Streamlit UI
 st.title("Christmas Tree Classifier")
@@ -28,17 +19,15 @@ if uploaded_file is not None:
     st.image(image, caption="Uploaded Image.", use_column_width=True)
 
     # Preprocess the image
-    image = image.resize((150, 150))  # Resize to match your model's input shape
-    image_array = np.array(image) / 255.0  # Normalize
-    image_array = np.expand_dims(image_array, axis=0)  # Add batch dimension
+    inputs = feature_extractor(images=image, return_tensors="tf")
 
     # Make prediction
-    prediction = model.predict(image_array)
-
-    # Interpret prediction (assuming binary classification)
-    class_names = ['Undecorated', 'Decorated']  # Update with your actual class names
-    predicted_class_index = 1 if prediction[0][0] >= 0.5 else 0  # Adjust threshold if needed
-    predicted_class = class_names[predicted_class_index]
+    logits = model(**inputs).logits
+    predicted_class_idx = tf.math.argmax(logits, axis=-1)[0]
+    
+    # Map class index to label
+    class_names = model.config.id2label  # Get class names from model config
+    predicted_class = class_names[predicted_class_idx]
 
     # Display the prediction
     st.write(f"Prediction: **{predicted_class}**")