Malware Detection LightGBM

LightGBM-based static malware detector for PE files.

Performance (hold-out test set)

  • AUC: 0.9978
  • Accuracy: 0.9895
  • Confusion matrix: [[4158, 66], [39, 5774]]

Artifacts

  • production_model.joblib
  • preprocessing_pipeline.joblib
  • feature_names.json

Notes

  • This repository contains model artifacts only.
  • For large CSV batch inference, use the Render app: https://malware-detection-ml-mihai.onrender.com/upload
Downloads last month
-
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Space using mihai-chindris/malware-detection-lgbm 1

Collection including mihai-chindris/malware-detection-lgbm

Evaluation results

  • AUC on Brazilian Malware Dataset (hold-out test set)
    self-reported
    0.998
  • Accuracy on Brazilian Malware Dataset (hold-out test set)
    self-reported
    0.990