Energy#

class scio.scores.Energy(*, act_norm=None, mode='raw', temperature=1.0)[source]#

Bases: BaseScoreClassif

Energy for classification.

Parameters:
  • temperature (float) – Temperature scaling factor. Defaults to 1.0.

  • mode – See BaseScoreClassif.

  • act_norm – See BaseScore.

References

[LWOL20]

Weitang Liu, Xiaoyun Wang, John Owens, and Yixuan Li. Energy-based Out-of-Distribution detection. In Advances in Neural Information Processing Systems, volume 33, 21464–21475. 2020. URL: https://proceedings.nips.cc/paper_files/paper/2020/file/f5496252609c43eb8a3d147ab9b9c006-Paper.pdf.