STEPDConfig

STEPDConfig#

class frouros.detectors.concept_drift.streaming.window_based.STEPDConfig(alpha_d: float = 0.003, alpha_w: float = 0.05, min_num_instances: int = 30)#

STEPD (Statistical test of equal proportions) [nishida2007detecting] configuration.

Parameters:

  • alpha_d (float) – significance value for overall, defaults to 0.003

  • alpha_w (float) – significance value for last, defaults to 0.05

  • min_num_instances (int) – minimum numbers of instances to start looking for changes, defaults to 30

References:

[nishida2007detecting]

Nishida, Kyosuke, and Koichiro Yamauchi. “Detecting concept drift using statistical testing.” Discovery science. Vol. 4755. 2007.

property alpha_d: float#

Significance level d property.

Returns:

significance level d

Return type:

float

property alpha_w: float#

Significance level w property.

Returns:

significance level w

Return type:

float

property min_num_instances: int#

Minimum number of instances property.

Returns:

minimum number of instances to start looking for changes

Return type:

int

Contents