1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
/* eslint no-console: 0, global-require: 0 */'use strict';var Benchmark = require('benchmark');var markdownTable = require('markdown-table');global.ss = require('simple-statistics');global.ssMaster = require('../');global.science = require('science');global.jStat = require('jstat').jStat;global.mathjs = require('mathjs');global.d3 = require('d3-array');
// We have a very simple way to benchmark single-array-input// methods, but jStat is a little more complicated - you specify// if you want _sample_ variance by adding a boolean argument to the// method. So I add it here.global.jStat.sampleVariance = function(input) {    return global.jStat.variance(input, true);};
function setupInput() {    var i;    var input10 = [];    for (i = 0; i < 10; i++) {        input10.push(Math.random());    }    var input100 = [];    for (i = 0; i < 100; i++) {        input100.push(Math.random());    }    var input1000 = [];    for (i = 0; i < 1000; i++) {        input1000.push(Math.random());    }
    var input10a = [];    for (i = 0; i < 10; i++) {        input10a.push(Math.random());    }    var input100a = [];    for (i = 0; i < 100; i++) {        input100a.push(Math.random());    }    var input1000a = [];    for (i = 0; i < 1000; i++) {        input1000a.push(Math.random());    }}
// This is the start of what will eventually be a markdown table// included in the readme.var results = [[    '', 'ss', 'science', 'jStat', 'mathjs', 'd3', 'ssMaster']];
// This is a generic 'suite maker' method that we use to test different// libraries doing the same work - it gets a name as a first argument, like// 'median', and the second argument is an array of strings that are the names// of the implementations of that function in different libraries.function suiteForMethods(name, methods, type) {
    var suite = new Benchmark.Suite({        name: name,        onError: function(error) {            console.log(error);        }    });
    methods.forEach(function (method) {        if (type === 'two-arrays') {            suite.add(method, {                setup: setupInput,                fn: method + '(input10, input10a);' + method + '(input100, input100a);' +                    method + '(input1000, input1000a);'            });        } else {            suite.add(method, {                setup: setupInput,                fn: method + '(input10);' + method + '(input100);' + method + '(input1000);'            });        }    });
    suite        .on('cycle', function(event) {            console.error(String(event.target));        })        .on('complete', function() {        var row = [name];        for (var i = 1; i < results[0].length; i++) {            row[i] = '';        }        var fastest = this.filter('fastest')[0];        this.forEach(function (event) {            // This gross code figures out which column to put            // each result in, because we don't have implementations            // of every method in every library.            var idx = results[0].indexOf(event.name.split('.')[1]);            row[idx] = Math.round(event.hz).toLocaleString();            if (event === fastest) {                row[idx] = '**' + row[idx] + '**';            }        });        results.push(row);    })    .run();}
// Here we declare all of the basic operations that we'll test, and the// functions in each library that do them.var tests = [    [        'variance',        [            'global.ss.variance',            'global.ssMaster.variance',            'global.jStat.variance'        ]    ],    [        'sampleVariance',        [            'global.ss.sampleVariance',            'global.ssMaster.sampleVariance',            'global.science.stats.variance',            'global.jStat.sampleVariance',            'global.d3.variance'        ]    ],    [        'covariance',        [            'global.ss.sampleCovariance',            'global.ssMaster.sampleCovariance',            'global.jStat.covariance'        ],        'two-arrays'    ],    [        'skewness',        [            'global.ss.sampleSkewness',            'global.ssMaster.sampleSkewness',            'global.jStat.skewness'        ]    ],    [        'geometricMean',        [            'global.ss.geometricMean',            'global.ssMaster.geometricMean',            'global.jStat.geomean'        ]    ],    [        'medianAbsoluteDeviation',        [            'global.ss.medianAbsoluteDeviation',            'global.ssMaster.medianAbsoluteDeviation',            'global.mathjs.mad',            'global.jStat.meddev'        ]    ],    [        'median',        [            'global.ss.median',            'global.ssMaster.median',            'global.mathjs.median',            'global.jStat.median',            'global.science.stats.median',            'global.d3.median'        ]    ],    [        'mode',        [            'global.ss.mode',            'global.ssMaster.mode',            'global.mathjs.mode',            'global.jStat.mode',            'global.science.stats.mode'        ]    ],    [        'mean',        [            'global.ss.mean',            'global.ssMaster.mean',            'global.mathjs.mean',            'global.jStat.mean',            'global.science.stats.mode'        ]    ],    [        'min',        [            'global.ss.min',            'global.ssMaster.min',            'global.mathjs.min',            'global.jStat.min',            'global.d3.min'        ]    ],    [        'max',        [            'global.ss.max',            'global.ssMaster.max',            'global.mathjs.max',            'global.jStat.max',            'global.d3.max'        ]    ],    [        'sum',        [            'global.ss.sum',            'global.ssMaster.sum',            'global.mathjs.sum',            'global.jStat.sum',            'global.d3.sum'        ]    ]];
tests.forEach(function (test) {    suiteForMethods(test[0], test[1], test[2]);});
// Finally, console.log spits out the content of readme.mdconsole.log('# Benchmarks');console.log('\n\nThis document is generated by `npm start`. It compares simple-statistics performance against similar libraries in JavaScript.');console.log('\n\n' + markdownTable(results));console.log('\n\nFastest libraries are in bold. Numbers are in operations per second.');console.log('\n\nssMaster is simple-statistics master branch, used for testing and proofing - it is not guaranteed to be any released version.');
console.log('\n\n### Versions\n');
['simple-statistics', 'science', 'jstat', 'mathjs', 'd3-array'].forEach(function (p) {    console.log('* ' + p + ': ' + require(p + '/package').version);});