streamsim.src.detectors.simple module

Simple Statistical Change Point Detector

This module provides a lightweight streaming change point detector that identifies significant deviations from recent signal behavior using a z-score-based approach.

Detection Method:

The detector maintains a rolling history of recent samples and computes the mean and standard deviation. A new sample is flagged as a change point if it deviates from the recent mean by more than a configurable number of standard deviations (threshold).

Important Dependencies:

• collections.deque: Efficient circular buffer

• streamsim.src.core.interfaces.StreamingChangePointDetector: Base interface

Author: F.Feenstra

class streamsim.src.detectors.simple.SimpleDetector(threshold: float = 2.0)

Bases: StreamingChangePointDetector

Very simple change point detector that flags points deviating from recent mean by more than a specified number of standard deviations.

update(x: float) → bool

Process a new sample and determine if it represents a change point.

Adds the sample to the rolling history, then checks if it deviates significantly from the recent mean using a z-score-based approach.

Parameters:

x (float) – The new sample value to evaluate.

Returns:

True if the sample is flagged as a change point, False otherwise.

Return type:

bool