meteorhacks/node-histo-utils

Name: node-histo-utils

Owner: meteorhacks

Description: A set of utilities to create, merge and manage histograms

Created: 2015-05-05 04:40:31.0

Updated: 2015-05-11 12:06:57.0

Pushed: 2015-05-11 12:07:15.0

Homepage: null

Size: 160

Language: JavaScript

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

histo-utils

A set of utilities to create, merge and manage histograms.

Installaton

i --save histo-utils

Building a histogram

HistoUtil = require('histo-utils');
histoBuilder = HistoUtils.create();
oBuilder.addPoints([10, 20]);
oBuilder.addPoints([30, 40, 50]);

uild the histogram
histogram = histoBuilder.build();
ole.log(histogram);
utput => { bins: [ [ 10, 3 ], [ 40, 2 ] ], binSize: 24 }

Merging Histograms

Merging Histograms. This don't gives us perfect histogram compared with a one generated with rawData. But still, this works pretty well if you can't store raw data.

histogramOne = { 
bins: [ [ 10, 3 ], [ 40, 2 ] ], 
binSize: 24 

histogramTwo = { 
bins: [ [ 10, 4 ], [ 50, 2 ] ], 
binSize: 32 


merged = HistoUtils.merge([histogramOne, histogramTwo]);
ole.log(merged);
utput => { bins: [ [ 22, 7 ], [ 52, 4 ] ], binSize: 27 }

Calculating Percentiles

Currently, percentile is the starting point of the bin. But, we can improve this once we stored the skewness for each bin along with the frequency.

histogram = { 
bins: [ [ 10, 30 ], [20, 40], [ 40, 20 ], [80, 20], [99999, 1] ], 
binSize: 24 


percentiles = HistoUtils.getPercentiles(histogram, [50, 95, 99]);
ole.log(percentiles);
utput => { '50': 20, '95': 80, '99': 80 }