Skip to content

CptNonPar 0.3.1

Latest

Choose a tag to compare

@EuanMcGonigle EuanMcGonigle released this 25 Nov 14:03
  • By default, the data is now centered and scaled before change point detection is applied, which improves empirical performance. This can be turned off by setting the new argument scale.data = FALSE in the functions np.mojo(), np.mojo.multilag(), and multiscale.np.mojo().