Table 12 Classification report of raw data after adversarial training under FGSM attack algorithm.
From: An incremental adversarial training method enables timeliness and rapid new knowledge acquisition
| Â | Precision | Recall | F1-score | support |
|---|---|---|---|---|
0 | 0.98571 | 0.98571 | 0.98571 | 140 |
1 | 1.00000 | 1.00000 | 1.00000 | 142 |
2 | 0.99383 | 0.98171 | 0.98773 | 164 |
3 | 1.00000 | 1.00000 | 1.00000 | 166 |
4 | 0.98571 | 1.00000 | 0.99281 | 138 |
Accuracy | Â | Â | 0.99333 | 750 |
Macro avg. | 0.99305 | 0.99348 | 0.99325 | 750 |
Weighted avg. | 0.99335 | 0.99333 | 0.99333 | 750 |
