Update app.py

app.py

@@ -1,29 +1,32 @@
-import numpy as np
 import streamlit as st
+import tensorflow as tf
+import numpy as np
 from PIL import Image
-import io
+import requests
+from io import BytesIO
 
-# Load the saved model
-from huggingface_hub import from_pretrained_keras
+# Download the model file from the Hugging Face repository
+model_url = "https://huggingface.co/Hammad712/Emotion_Detection/resolve/main/emotion_detection_model.h5"
+response = requests.get(model_url)
+model_file_path = 'emotion_detection_model.h5'
 
-model = from_pretrained_keras("Hammad712/Emotion_Detection")
+# Save the downloaded model file
+with open(model_file_path, 'wb') as f:
+    f.write(response.content)
+
+# Load the saved model
+model = tf.keras.models.load_model(model_file_path)
 
 # Define the emotion labels
 emotion_labels = ['happy', 'sad', 'neutral']
 
 # Function to preprocess the image for the model
 def preprocess_image(image):
-    # Resize the image to match the input size of the model
     image = image.resize((48, 48))
-    # Convert the image to grayscale
     image = image.convert('L')
-    # Convert the image to a numpy array
     image = np.array(image)
-    # Normalize the image
     image = image / 255.0
-    # Expand the dimensions to match the input shape of the model
     image = np.expand_dims(image, axis=-1)
-    # Expand the dimensions to create a batch of size 1
     image = np.expand_dims(image, axis=0)
     return image
 
@@ -90,7 +93,7 @@ if image_file is not None:
         st.markdown(f'<div class="predicted-emotion">Predicted Emotion: {predicted_label}</div>', unsafe_allow_html=True)
 
         # Save the prediction and image to session state
-        image_bytes = io.BytesIO()
+        image_bytes = BytesIO()
         image.save(image_bytes, format='PNG')
         st.session_state.predictions.append({
             'image': image_bytes.getvalue(),