skada.metrics.SoftNeighborhoodDensity

class skada.metrics.SoftNeighborhoodDensity(T=0.05, greater_is_better=True, kwargs=None)

Score based on the entropy of similarity between unsupervised dataset.

See [19] for details.

Parameters:

Tfloat

Temperature in the Eq. 2 in [1]_. Default is set to 0.05, the value proposed in the paper.

greater_is_betterbool, default=True

Whether scorer is a score function (default), meaning high is good, or a loss function, meaning low is good. In the latter case, the scorer object will sign-flip the outcome of the scorer.

References

[19]

Kuniaki Saito et al. Tune it the Right Way: Unsupervised Validation of Domain Adaptation via Soft Neighborhood Density. International Conference on Computer Vision, 2021.