GCN · full_fcg
Standard SplitFeb 17, 2026ee3c8434946241af93299ac7129d15cb
Description
Train GCN on the full_fcg dataset to complete the GCN comparison across both graph representations.
Conclusion
Similar to GCN on internal_only (94.4% vs 96.3% accuracy). The full_fcg representation provides no clear benefit for GCN either. Recall remains at 93.8%.
Test Metrics
Accuracy
94.4%
F1 Macro
93.5%
F1 Malware
90.9%
Precision
88.2%
Recall
93.8%
AUROC
98.9%
Best Val Loss
0.1278
Training Time
1449.2000s
Confusion Matrix
| Pred Benign | Pred Malware | |
|---|---|---|
| Actual Benign | 72 | 4 |
| Actual Malware | 2 | 30 |
Configuration
| Hidden Dim | 128 |
| Num Layers | 3 |
| Dropout | 0.5 |
| Batch Size | 4 |
| Learning Rate | 0.001 |
| Weight Decay | 0.0001 |
| Max Epochs | 200 |
| ES Patience | 20 |
| ES Min Epochs | 100 |
| LR Patience | 10 |
| LR Factor | 0.5 |
| Mixed Precision | Yes |
| Random Seed | 42 |
| Epochs Trained | 101 |