algorithms.change_point

Module: algorithms.change_point

Inheritance diagram for selectinf.algorithms.change_point:

digraph inheritance4ce2426b7e { rankdir=LR; size="8.0, 12.0"; "algorithms.change_point.change_point" [URL="#selectinf.algorithms.change_point.change_point",fontname="Vera Sans, DejaVu Sans, Liberation Sans, Arial, Helvetica, sans",fontsize=10,height=0.25,shape=box,style="setlinewidth(0.5)",target="_top",tooltip="Estimate change points using multiscale changepoint "]; }

change_point

class selectinf.algorithms.change_point.change_point(Y, minsize=None)

Bases: object

Estimate change points using multiscale changepoint penatly.

__init__(Y, minsize=None)

Initialize self. See help(type(self)) for accurate signature.

ndraw = 50

quantile = 0.95

sigma = None

fit(solve_args={'min_its': 200})

merge_intervals(intervals)

cluster the endpoints into intervals by saying points that are M.minsize apart or less are in the same cluster

selectinf.algorithms.change_point.one_jump_instance(delta, p=60, sigma=1)

Data generating mechanism of Figure 1 in [http://arxiv.org/abs/1606.03552](http://arxiv.org/abs/1606.03552).

Parameters

delta : float

Signal size

p : int

Shape of signal.

sigma : float

Noise variance – both signal and noise are scaled by this scalar.