Create app.py

app.py

@@ -0,0 +1,36 @@
+import gradio as gr
+from tensorflow.keras.models import load_model
+import numpy as np
+
+# Load the model
+model = load_model(r"C:\Users\uSeR\Documents\ai-tutor-ruangguru\project\waste_classifier_model.keras")
+
+# Prediction function
+def classify_image(image):
+    # Ensure the image is in the expected format
+    if image is None:
+        return "No image provided."
+
+    # Convert the image to a numpy array
+    image = np.array(image)
+
+    # Check if the image has the expected shape and resize if necessary
+    if image.shape != (128, 128, 3):
+        image = np.resize(image, (128, 128, 3))  # Resize the image to (128, 128, 3)
+
+    # Normalize the image
+    image = image / 255.0
+
+    # Add batch dimension
+    image = np.expand_dims(image, axis=0)
+
+    # Make prediction
+    prediction = model.predict(image)
+
+    # Interpret the prediction
+    class_label = 'Organic' if prediction[0][0] > 0.5 else 'Recycleable'  # Adjust labels based on your model's output
+    return class_label
+
+# Create Gradio interface
+interface = gr.Interface(fn=classify_image, inputs=gr.Image(type="numpy"), outputs="text")
+interface.launch(share=True)