initial commit

app.py

@@ -0,0 +1,82 @@
+import gradio as gr
+import numpy as np
+from PIL import Image
+import tensorflow as tf
+import logging
+import os
+
+# Set up logging to capture errors
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+logger.info("Starting application initialization")
+
+# Load class names from class_indices.txt
+try:
+    logger.info("Loading class names from class_indices.txt")
+    with open("class_indices.txt", "r") as f:
+        class_names = [line.strip() for line in f.readlines() if line.strip()]
+    logger.info(f"Class names loaded: {class_names}")
+except Exception as e:
+    logger.error(f"Failed to load class_indices.txt: {str(e)}")
+    raise Exception(f"Failed to load class_indices.txt: {str(e)}")
+
+# Load the Keras .h5 model
+try:
+    logger.info("Loading model: garbage_classifier.h5")
+    model = tf.keras.models.load_model("garbage_classifier.h5", compile=False)
+    logger.info("Model loaded successfully")
+except Exception as e:
+    logger.error(f"Failed to load model: {str(e)}")
+    raise Exception(f"Failed to load model: {str(e)}")
+
+def predict_image(image: Image.Image):
+    try:
+        logger.info("Processing image for prediction")
+        # Preprocess the image
+        img = image.convert("RGB").resize((128, 128))
+        arr = np.array(img).astype("float32") / 255.0
+        arr = np.expand_dims(arr, axis=0)
+
+        # Make prediction
+        preds = model.predict(arr)[0]
+        logger.info("Prediction completed")
+
+        # Get predicted class and confidence
+        pred_class_idx = np.argmax(preds)
+        pred_class = class_names[pred_class_idx]
+        confidence = float(preds[pred_class_idx])
+
+        # Format per-class probabilities
+        prob_text = "\n".join([f"{class_names[i]}: {float(preds[i]):.2f}" for i in range(len(class_names))])
+
+        # Return formatted output
+        output = f"Prediction\nClass: {pred_class}\nConfidence: {confidence:.2f}\n\nPer-class probabilities\n{prob_text}"
+        return output
+    except Exception as e:
+        logger.error(f"Error during prediction: {str(e)}")
+        return f"Error during prediction: {str(e)}"
+
+# Create Gradio interface
+logger.info("Initializing Gradio interface")
+with gr.Blocks() as iface:
+    gr.Markdown("# ♻️ Garbage Classification Application")
+    gr.Markdown("Upload an image of garbage (plastic, organic, or metal):")
+    img_input = gr.Image(type="pil")
+    output = gr.Textbox(label="Prediction Result")
+    gr.Button("Classify").click(
+        fn=predict_image,
+        inputs=img_input,
+        outputs=output
+    )
+
+# Queue the interface for serving (important for Spaces)
+iface.queue(api_open=False)  # Disable public API for simplicity
+
+logger.info("Gradio interface queued and ready for serving")
+
+# Block to keep the process alive (Spaces handles serving)
+if __name__ == "__main__":
+    logger.info("Entering blocking mode for Spaces serving")
+    while True:
+        pass

class_indices.txt

@@ -0,0 +1,3 @@
+metal
+organic
+plastic

garbage_classifier.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4939524ee92a9e43cf0f73e0bf2f5ffda76f626a037e0fc818c1daf755af811c
+size 26340920

requirements.txt

@@ -0,0 +1,4 @@
+gradio==5.46.0
+numpy==1.26.4
+Pillow==10.3.0
+tensorflow==2.16.1