Module org.elasticsearch.server

Package org.elasticsearch.index.codec.vectors.cluster

Class OnlineQuantileEstimator

java.lang.Object

org.elasticsearch.index.codec.vectors.cluster.OnlineQuantileEstimator

public class OnlineQuantileEstimator
extends Object

Computes an approximation of the q-quantile of a data stream using constant memory. It uses this approach that was originally designed for integers. This implementation works with floating-point numbers by quantizing the stream values. This requires knowing the minimum and maximum values in the stream and choosing a desired precision. Note that the precision only pertains to the quantization and the quantile estimator carries additional error.

Method Summary

Modifier and Type	Method	Description
float	getEstimate()	Return the q-quantile estimate.
void	updateEstimate(float value)	Update the estimate with a new value from the stream.
void	updateEstimate(float[] values)	Update the estimate with a batch of values from the stream.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Details

updateEstimate

public void updateEstimate(float value)

Update the estimate with a new value from the stream.

Parameters:

value - the new value from the stream

updateEstimate

public void updateEstimate(float[] values)

Update the estimate with a batch of values from the stream.

Parameters:

values - the new values from the stream

getEstimate

public float getEstimate()

Return the q-quantile estimate.