Update app.py

app.py

@@ -9,14 +9,19 @@ from tensorflow.keras.models import load_model
 from tensorflow.keras.preprocessing import image
 import numpy as np
 
-# Load the pre-trained model
-new_model = load_model('cat_classifier_model.h5')
+# Load the trained model
+model = load_model('/content/cat_classifier_model.h5')
 
-def classify_image(image_path):
-    img = image.load_img(image_path, target_size=(224, 224))
-    img_array = image.img_to_array(img)
+# Function to predict whether an image contains a cat
+def predict_cat(image_content):
+    # Convert image content to PIL Image
+    img = Image.open(BytesIO(image_content))
+    img = img.convert('RGB')
+    img = img.resize((224, 224))
+
+    img_array = np.array(img)
     img_array = np.expand_dims(img_array, axis=0)
-    img_array /= 255.0  # Rescale to values between 0 and 1 (same as during training)
+    img_array = img_array / 255.0  # Rescale to values between 0 and 1 (same as during training)
 
     prediction = model.predict(img_array)
     if prediction[0][0] > 0.5:
@@ -26,15 +31,13 @@ def classify_image(image_path):
 
 # Create a Gradio interface
 iface = gr.Interface(
-    fn=classify_image,
-    inputs=gr.Image(),
-    outputs="text",
-    live=True,
-      
+    fn=predict_cat,
+    inputs=gr.Image(type='file', label='Upload an image of a tablet'),
+    outputs='text'
 )
 
-# Launch the Gradio interface
-iface.launch()
+# Launch the interface with share=True to create a public link
+iface.launch(share=True)