v1.2.0

v1.2.0 by @elsacho: Added new proper loss functions:

• ProperLpLoss(p=p): Metrics to evaluate $E[ \Vert f(X) - E[Y|f(X)] \Vert_p ]$ where $f(X)$ are the
  predictions of the classifier, $p >= 1$, including p=float("inf")
• TopClassLoss: A wrapper to variationally evaluate top-class errors.
• OverConfidenceLoss & UnderConfidenceLoss: Wrappers to variationally evaluate
  over/under-confidence in binary predictors.
• MetricsWithCalibration can now handle arbitrary classifiers and Lp-type losses.
• New classifiers: Added WS_CatboostClassifier and WS_LGBMClassifier for
  evaluating calibration errors.
• removed sklearn < 1.7 constraint.

v1.1.0

v1.1.0 by @eugeneberta: Improvements to the SVS and SMS calibrators:

• logit pre-processing with 'ts-mix' is now automatic,
  and the global scaling parameter $\alpha$ is fixed to 1. This yields:
  • improved performance on our tabular and computer vision benchmarks
    (see the arxiv v2 of the SMS paper, coming soon).
  • faster convergence.
  • ability to compute the duality gap in closed form for stopping SAGA solvers,
    which we implement in this version.
• improved L-BFGS solvers, much faster than in the previous version.
  Now the solver for default SVS and SMS.
• the default binary calibrator in LogisticCalibrator is now quadratic scaling
  instead of affine scaling, this can be changed back by using
  LogisticCalibrator(binary_type='affine').

v1.0.0

What's Changed

• New post-hoc calibrators including SMS, SVS, affine and quadratic scaling. by @eugeneberta in #1

New Contributors

• @eugeneberta made their first contribution in #1

Full Changelog: v0.0.2...v1.0.0