Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
<!DOCTYPE html>
|
| 2 |
+
<html lang="en">
|
| 3 |
+
<head>
|
| 4 |
+
<meta charset="UTF-8">
|
| 5 |
+
<meta name="viewport" content="width=device-width, initial-scale=1">
|
| 6 |
+
<title>Garbage Classification Model — CS549</title>
|
| 7 |
+
<style>
|
| 8 |
+
body { font-family: Arial, sans-serif; margin: 40px; line-height: 1.6; color: #333; }
|
| 9 |
+
h1 { color: #2c3e50; }
|
| 10 |
+
h2 { color: #34495e; }
|
| 11 |
+
code { background: #f4f4f4; padding: 2px 4px; border-radius: 4px; }
|
| 12 |
+
pre { background: #f9f9f9; padding: 10px; border-left: 3px solid #ccc; }
|
| 13 |
+
ul { margin: 0; padding-left: 20px; }
|
| 14 |
+
</style>
|
| 15 |
+
</head>
|
| 16 |
+
<body>
|
| 17 |
+
|
| 18 |
+
<h1>🗑️ Garbage Classification Model — CS549</h1>
|
| 19 |
+
|
| 20 |
+
<p>A Convolutional Neural Network trained to classify garbage images into 7 categories:</p>
|
| 21 |
+
<ul>
|
| 22 |
+
<li>Cardboard</li>
|
| 23 |
+
<li>Glass</li>
|
| 24 |
+
<li>Metal</li>
|
| 25 |
+
<li>Paper</li>
|
| 26 |
+
<li>Plastic</li>
|
| 27 |
+
<li>Trash</li>
|
| 28 |
+
<li>Biodegradable</li>
|
| 29 |
+
</ul>
|
| 30 |
+
|
| 31 |
+
<h2>🔍 Use Case</h2>
|
| 32 |
+
<p>Automates waste sorting to improve recycling and support eco-friendly efforts. Ideal for educational demos, PoC apps, or smart bins.</p>
|
| 33 |
+
|
| 34 |
+
<h2>🧠 Model Overview</h2>
|
| 35 |
+
<ul>
|
| 36 |
+
<li><strong>Architecture:</strong> CNN (Conv2D, MaxPooling, Dense)</li>
|
| 37 |
+
<li><strong>Input:</strong> 224x224 RGB</li>
|
| 38 |
+
<li><strong>Output:</strong> Probabilities for 7 classes</li>
|
| 39 |
+
<li><strong>Accuracy:</strong> ~92% on validation set</li>
|
| 40 |
+
<li><strong>Extras:</strong> Grad-CAM for explainability</li>
|
| 41 |
+
</ul>
|
| 42 |
+
|
| 43 |
+
<h2>📦 How to Use</h2>
|
| 44 |
+
<pre><code>from tensorflow.keras.models import load_model
|
| 45 |
+
from PIL import Image
|
| 46 |
+
import numpy as np
|
| 47 |
+
|
| 48 |
+
model = load_model("GarbageMLModel_CS549.h5")
|
| 49 |
+
|
| 50 |
+
img = Image.open("example.jpg").resize((224, 224))
|
| 51 |
+
img_array = np.expand_dims(np.array(img) / 255.0, axis=0)
|
| 52 |
+
|
| 53 |
+
pred = model.predict(img_array)
|
| 54 |
+
classes = ['cardboard', 'glass', 'metal', 'paper', 'plastic', 'trash', 'biodegradable']
|
| 55 |
+
print("Prediction:", classes[np.argmax(pred)])
|
| 56 |
+
</code></pre>
|
| 57 |
+
|
| 58 |
+
<h2>📁 Files</h2>
|
| 59 |
+
<ul>
|
| 60 |
+
<li><code>GarbageMLModel_CS549.h5</code> – Trained Keras model</li>
|
| 61 |
+
<li><code>label_map.json</code> – Class labels</li>
|
| 62 |
+
</ul>
|
| 63 |
+
|
| 64 |
+
<h2>👤 Author</h2>
|
| 65 |
+
<p>Vincent Huynh<br>
|
| 66 |
+
📧 <a href="mailto:vintendohuynh@gmail.com">vintendohuynh@gmail.com</a><br>
|
| 67 |
+
🔗 <a href="#">LinkedIn</a> | 🔗 <a href="#">GitHub</a></p>
|
| 68 |
+
|
| 69 |
+
</body>
|
| 70 |
+
</html>
|