update readme
Browse files
Readme.md
CHANGED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
### Architecture Model
|
| 2 |
+
The model is based on ResNet18 as a backbone and fine-tuning only the last convolutional block (layer4). It also use custom classifier for Hidden layer of 256 units with ReLU activation,
|
| 3 |
+
Dropout layer (0.3) for regularization, and Output layer with 6 units (one per waste category).
|
| 4 |
+
|
| 5 |
+
### Model Performance
|
| 6 |
+
The model achieves strong performance across all waste categories, with an overall accuracy of 96%. Detailed performance metrics per class:
|
| 7 |
+
| Category | Precision | Recall | F1-Score | Support |
|
| 8 |
+
|------------|-----------|--------|----------|---------|
|
| 9 |
+
| Cardboard | 0.99 | 0.95 | 0.97 | 161 |
|
| 10 |
+
| Glass | 0.92 | 0.98 | 0.95 | 200 |
|
| 11 |
+
| Metal | 0.97 | 0.93 | 0.95 | 164 |
|
| 12 |
+
| Paper | 0.97 | 0.96 | 0.96 | 238 |
|
| 13 |
+
| Plastic | 0.96 | 0.95 | 0.96 | 193 |
|
| 14 |
+
| Trash | 0.90 | 0.96 | 0.93 | 55 |
|