Update app.py

app.py

@@ -1,52 +1,40 @@
 import gradio as gr
 import tensorflow as tf
 import numpy as np
-from PIL import Image
-from io import BytesIO
 
-# Define image dimensions
-img_height = 224  # Replace with the actual height of your images
-img_width = 224   # Replace with the actual width of your images
+# Load your trained TensorFlow model
+model = tf.keras.models.load_model('best_model_weights.h5')  # Load your saved model
 
-# Load your trained model
-model = tf.keras.models.load_model("best_model_weights.h5")  # Replace with the path to your saved model
-
-# Define the image classification function
+# Define a function to make predictions
 def classify_image(input_image):
-    try:
-        # Preprocess the input image
-        input_image = Image.open(BytesIO(input_image))
-        input_image = input_image.resize((img_width, img_height))
-        input_image = np.array(input_image) / 255.0  # Normalize pixel values
-
-        # Make a prediction using the model
-        predictions = model.predict(np.expand_dims(input_image, axis=0))
-
-        # Get the class label with the highest probability
-        class_index = np.argmax(predictions)
-        class_prob = predictions[0][class_index]
-
-        # Define class labels (you can replace these with your actual class labels)
-        class_labels = ["Normal", "Cataract"]
+    # Preprocess the input image (resize and normalize)
+    input_image = tf.image.resize(input_image, (224, 224))  # Make sure to match your model's input size
+    input_image = (input_image / 255.0)  # Normalize to [0, 1]
+    input_image = np.expand_dims(input_image, axis=0)  # Add batch dimension
 
-        # Get the class label
-        class_label = class_labels[class_index]
+    # Make a prediction using your model
+    prediction = model.predict(input_image)
 
-        return f"Predicted Class: {class_label} (Probability: {class_prob:.2f})"
-    except Exception as e:
-        return "Error: Invalid image file"
+    # Assuming your model outputs probabilities for two classes, you can return the class with the highest probability
+    class_index = np.argmax(prediction)
+    class_labels = ["Class 0", "Class 1"]  # Replace with your actual class labels
+    predicted_class = class_labels[class_index]
 
-# ...
+    return predicted_class
 
+# Create a Gradio interface
+input_interface = gr.inputs.Image()  # Gradio input component for image
+output_interface = gr.outputs.Text()  # Gradio output component for text
 
-# Define the Gradio interface
-iface = gr.Interface(
+# Create the Gradio app
+app = gr.Interface(
     fn=classify_image,
-    inputs=gr.inputs.Image(shape=(img_height, img_width)),
-    outputs="text",
+    inputs=input_interface,
+    outputs=output_interface,
     live=True,
-    title="Image Classifier"
+    title="Image Classifier",
+    description="Classify images using a trained model."
 )
 
-# Run the Gradio interface
-iface.launch()
+# Start the Gradio app
+app.launch()