Shortcuts

ThresholdAccuracy

class torch_uncertainty.metrics.regression.ThresholdAccuracy(power, lmbda=1.25, **kwargs)[source]

The Threshold Accuracy metric, a.k.a. d1, d2, d3.

Parameters:
  • power – The power to raise the threshold to. Often in [1, 2, 3].

  • lmbda – The threshold to compare the max of ratio of predictions to targets and its inverse to. Defaults to 1.25.

  • kwargs – Additional keyword arguments, see Advanced metric settings.

compute()[source]

Compute the Threshold Accuracy.

update(preds, target)[source]

Update state with predictions and targets.