All pull requests

Pull requests

• Return df in T.params and fix the Cauchy starting fit

  #450·alejandroschuler opened Oct 27, 2026Collaborator·

  ·

• Compute the LogitNormal mean by quadrature

  #449·alejandroschuler opened Oct 27, 2026Collaborator·

  ·

• Count NGBClassifier classes with Dist.n_classes (fixes BetaBernoulli after #413)

  #448·alejandroschuler opened Oct 27, 2026Collaborator·

  ·

• List all distributions in the user guide tables

  #447·alejandroschuler opened Oct 27, 2026Collaborator·

  ·

• Add a NegativeBinomial distribution built with the SymPy factory

  #446·alejandroschuler opened Oct 27, 2026Collaborator·

  ·

• Make SymPy factory distributions picklable

  #445·alejandroschuler opened Oct 27, 2026Collaborator·

  ·

• Fix the T distribution crash: cap df and use the exact Fisher information

  #443·alejandroschuler opened Oct 27, 2026Collaborator·

  ·
  2

• Support integer indexing of Categorical and MultivariateNormal predictions

  #442·alejandroschuler opened Oct 27, 2026Collaborator·

  ·

• Store params in NormalFixedVar and NormalFixedMean so they can be sliced

  #441·alejandroschuler opened Oct 27, 2026Collaborator·

  ·

• Add df to the params of the T distributions

  #440·alejandroschuler opened Oct 27, 2026Collaborator·

  ·

• Deprecate the var attribute of the Normal and T distributions

  #439·alejandroschuler opened Oct 27, 2026Collaborator·

  ·

• Name the MultivariateNormal covariance cov in params

  #438·alejandroschuler opened Oct 27, 2026Collaborator·

  ·

• Mark only the T case as xfail in the logscore distribution test

  #437·alejandroschuler opened Oct 27, 2026Collaborator·

  ·

• Stop overflow warnings from rejected line-search steps

  #434·alejandroschuler opened Oct 27, 2026Collaborator·

  ·
  1

• Add mu and sigma to LogNormal and raise AttributeError for unknown names

  #433·alejandroschuler opened Oct 27, 2026Collaborator·

  ·
  1

• Make NGBSurvival work with clone and GridSearchCV

  #432·alejandroschuler opened Oct 27, 2026Collaborator·

  ·
  1