├── .github
    └── workflows
    │   ├── documentationjs.yml
    │   ├── lactame.yml
    │   ├── nodejs.yml
    │   └── release.yml
├── .gitignore
├── .npmrc
├── .prettierrc.json
├── CHANGELOG.md
├── CODE_OF_CONDUCT.md
├── CONTRIBUTING.md
├── LICENSE
├── README.md
├── babel.config.json
├── bower.json
├── eslint.config.mjs
├── examples
    ├── array.js
    ├── fcnnls.js
    ├── leafDataset
    │   ├── README.md
    │   ├── knn.js
    │   ├── leaf.csv
    │   ├── logreg.js
    │   ├── naive-bayes.js
    │   ├── package.json
    │   └── svm.js
    └── sg.js
├── package.json
├── rollup.config.mjs
└── src
    ├── __tests__
        └── index.test.js
    └── index.js


/.github/workflows/documentationjs.yml:
--------------------------------------------------------------------------------
 1 | name: Deploy documentation.js on GitHub pages
 2 | 
 3 | on:
 4 |   workflow_dispatch:
 5 |   release:
 6 |     types: [published]
 7 | 
 8 | jobs:
 9 |   deploy:
10 |     runs-on: ubuntu-latest
11 |     steps:
12 |       - uses: actions/checkout@v2
13 |       - name: Build documentation
14 |         uses: zakodium/documentationjs-action@v1
15 |       - name: Deploy to GitHub pages
16 |         uses: JamesIves/github-pages-deploy-action@releases/v4
17 |         with:
18 |           token: ${{ secrets.BOT_TOKEN }}
19 |           branch: gh-pages
20 |           folder: docs
21 |           clean: true
22 | 


--------------------------------------------------------------------------------
/.github/workflows/lactame.yml:
--------------------------------------------------------------------------------
 1 | name: Deploy build on lactame.com
 2 | 
 3 | on:
 4 |   workflow_dispatch:
 5 |   release:
 6 |     types: [published]
 7 | 
 8 | env:
 9 |   NODE_VERSION: 14.x
10 | 
11 | jobs:
12 |   deploy:
13 |     runs-on: ubuntu-latest
14 |     steps:
15 |       - uses: actions/checkout@v2
16 |       - name: Get package name
17 |         run: echo "PACKAGE_NAME=$(jq .name package.json | tr -d '"')" >> $GITHUB_ENV
18 |       - uses: actions/setup-node@v2
19 |         with:
20 |           node-version: ${{ env.NODE_VERSION }}
21 |       - name: Install dependencies
22 |         run: npm install
23 |       - name: Build project
24 |         run: npm run build
25 |       - name: Deploy to lactame.com
26 |         uses: zakodium/lactame-action@v1
27 |         with:
28 |           token: ${{ secrets.LACTAME_TOKEN }}
29 |           name: ${{ env.PACKAGE_NAME }}
30 |           folder: dist
31 | 


--------------------------------------------------------------------------------
/.github/workflows/nodejs.yml:
--------------------------------------------------------------------------------
 1 | name: Node.js CI
 2 | 
 3 | on:
 4 |   push:
 5 |     branches:
 6 |       - main
 7 |   pull_request:
 8 | 
 9 | jobs:
10 |   nodejs:
11 |     # Documentation: https://github.com/zakodium/workflows#nodejs-ci
12 |     uses: zakodium/workflows/.github/workflows/nodejs.yml@nodejs-v1
13 |     with:
14 |       test-setup-command: 'npm run build && node dist/ml.js && node dist/ml.min.js'
15 | 


--------------------------------------------------------------------------------
/.github/workflows/release.yml:
--------------------------------------------------------------------------------
 1 | name: Release
 2 | 
 3 | on:
 4 |   push:
 5 |     branches:
 6 |       - main
 7 | 
 8 | jobs:
 9 |   release:
10 |     # Documentation: https://github.com/zakodium/workflows#release
11 |     uses: zakodium/workflows/.github/workflows/release.yml@release-v1
12 |     with:
13 |       npm: true
14 |     secrets:
15 |       github-token: ${{ secrets.BOT_TOKEN }}
16 |       npm-token: ${{ secrets.NPM_BOT_TOKEN }}
17 | 


--------------------------------------------------------------------------------
/.gitignore:
--------------------------------------------------------------------------------
1 | .idea
2 | node_modules
3 | lib
4 | 
5 | .eslintcache
6 | dist
7 | coverage


--------------------------------------------------------------------------------
/.npmrc:
--------------------------------------------------------------------------------
1 | package-lock=false
2 | 


--------------------------------------------------------------------------------
/.prettierrc.json:
--------------------------------------------------------------------------------
1 | {
2 |   "arrowParens": "always",
3 |   "semi": true,
4 |   "singleQuote": true,
5 |   "tabWidth": 2,
6 |   "trailingComma": "all"
7 | }
8 | 


--------------------------------------------------------------------------------
/CHANGELOG.md:
--------------------------------------------------------------------------------
  1 | # Changelog
  2 | 
  3 | ## [8.0.0](https://github.com/mljs/ml/compare/v7.0.0...v8.0.0) (2024-10-21)
  4 | 
  5 | 
  6 | ### ⚠ BREAKING CHANGES
  7 | 
  8 | * update dependencies ([#186](https://github.com/mljs/ml/issues/186))
  9 | 
 10 | ### Features
 11 | 
 12 | * update dependencies ([#186](https://github.com/mljs/ml/issues/186)) ([9cf0471](https://github.com/mljs/ml/commit/9cf0471c1c2787b136f6e1ad537fee20abc02a39))
 13 | 
 14 | 
 15 | ### Bug Fixes
 16 | 
 17 | * rename rollup to rollup.config.mjs ([c63c278](https://github.com/mljs/ml/commit/c63c2784f84c4909081ae0a2bf2e6b17619f75dd))
 18 | 
 19 | ## [7.0.0](https://github.com/mljs/ml/compare/v5.3.0...v7.0.0) (2024-02-02)
 20 | 
 21 | 
 22 | ### ⚠ BREAKING CHANGES
 23 | 
 24 | * update dependencies
 25 | * update dependencies
 26 | * remove support for Node.js 12
 27 | * remove num-sort package exportation
 28 | * update all dependencies
 29 | * Remove dist from git
 30 | 
 31 | ### release-as
 32 | 
 33 | * 7.0.0 ([27176ee](https://github.com/mljs/ml/commit/27176ee3eefa493144d76296ec3401c2c84fe190))
 34 | 
 35 | 
 36 | ### Features
 37 | 
 38 | * add lactame.com action release ([2547a9a](https://github.com/mljs/ml/commit/2547a9a1760582612e3c5a012f5ff977655a3f6c))
 39 | * Remove dist from git ([90392a5](https://github.com/mljs/ml/commit/90392a5dec7e1d75cf73398cdfffc0e91765302f))
 40 | * remove num-sort package exportation ([4b8c47b](https://github.com/mljs/ml/commit/4b8c47b25a903dd06d664afd84d3738bdcbc26ae))
 41 | * update all dependencies ([114d814](https://github.com/mljs/ml/commit/114d81448d47eca961223f2c2f38da297466c238))
 42 | * update and improve GSD ([7222a2c](https://github.com/mljs/ml/commit/7222a2c797dfff0ae1b4777dbbd09c3c41488187))
 43 | * update dependencies ([8adb8f1](https://github.com/mljs/ml/commit/8adb8f1ddb4842b32afe92a28a354408034889a3))
 44 | 
 45 | 
 46 | ### Bug Fixes
 47 | 
 48 | * add missing ml-array-sum package ([d471033](https://github.com/mljs/ml/commit/d471033dc121a50313a2ab8ae0cccdb61f26c113))
 49 | * correct link to lactame.com to v6.0.0 ([65e9ca0](https://github.com/mljs/ml/commit/65e9ca0629dfc6b64ea1b17e8c11273e97fa0656))
 50 | * update dependencies to fix nGMCA ([9b040d0](https://github.com/mljs/ml/commit/9b040d0fa66baca16f3c8fb2ebb2a74eb39f5b46))
 51 | 
 52 | 
 53 | ### Documentation
 54 | 
 55 | * update Array.seqentialFill Link ([#167](https://github.com/mljs/ml/issues/167)) ([46bb1ec](https://github.com/mljs/ml/commit/46bb1ec2150e496f67ca617acb2da2e2894dbfdc))
 56 | 
 57 | 
 58 | ### Miscellaneous Chores
 59 | 
 60 | * remove support for Node.js 12 ([88c0717](https://github.com/mljs/ml/commit/88c0717ffb4e5ee0b3cc0401bd7f38fb96c3eade))
 61 | * update dependencies ([88c0717](https://github.com/mljs/ml/commit/88c0717ffb4e5ee0b3cc0401bd7f38fb96c3eade))
 62 | 
 63 | ## [6.0.0](https://github.com/mljs/ml/compare/v5.3.0...v6.0.0) (2021-06-10)
 64 | 
 65 | 
 66 | ### ⚠ BREAKING CHANGES
 67 | 
 68 | * remove num-sort package exportation
 69 | * update all dependencies
 70 | * Remove dist from git
 71 | 
 72 | ### Features
 73 | 
 74 | * Remove dist from git ([90392a5](https://github.com/mljs/ml/commit/90392a5dec7e1d75cf73398cdfffc0e91765302f))
 75 | * remove num-sort package exportation ([4b8c47b](https://github.com/mljs/ml/commit/4b8c47b25a903dd06d664afd84d3738bdcbc26ae))
 76 | * update all dependencies ([114d814](https://github.com/mljs/ml/commit/114d81448d47eca961223f2c2f38da297466c238))
 77 | * update and improve GSD ([7222a2c](https://github.com/mljs/ml/commit/7222a2c797dfff0ae1b4777dbbd09c3c41488187))
 78 | 
 79 | 
 80 | ### Bug Fixes
 81 | 
 82 | * add missing ml-array-sum package ([d471033](https://github.com/mljs/ml/commit/d471033dc121a50313a2ab8ae0cccdb61f26c113))
 83 | * correct link to lactame.com to v6.0.0 ([65e9ca0](https://github.com/mljs/ml/commit/65e9ca0629dfc6b64ea1b17e8c11273e97fa0656))
 84 | * update dependencies to fix nGMCA ([9b040d0](https://github.com/mljs/ml/commit/9b040d0fa66baca16f3c8fb2ebb2a74eb39f5b46))
 85 | 
 86 | <a name="3.5.0"></a>
 87 | # [3.5.0](https://github.com/mljs/ml/compare/v3.4.0...v3.5.0) (2018-08-16)
 88 | 
 89 | 
 90 | ### Bug Fixes
 91 | 
 92 | * **package:** update ml-xsadd to version 2.0.0 ([#108](https://github.com/mljs/ml/issues/108)) ([01c5e25](https://github.com/mljs/ml/commit/01c5e25))
 93 | * require ml-random correctly ([4d64b91](https://github.com/mljs/ml/commit/4d64b91))
 94 | * require xsadd correctly ([facfaea](https://github.com/mljs/ml/commit/facfaea))
 95 | 
 96 | 
 97 | 
 98 | <a name="3.4.0"></a>
 99 | # [3.4.0](https://github.com/mljs/ml/compare/v3.3.0...v3.4.0) (2018-08-02)
100 | 
101 | 
102 | 
103 | <a name="3.3.0"></a>
104 | # [3.3.0](https://github.com/mljs/ml/compare/v3.2.0...v3.3.0) (2018-07-29)
105 | 
106 | 
107 | ### Bug Fixes
108 | 
109 | * DecisionTreeRegression import typo ([#97](https://github.com/mljs/ml/issues/97)) ([c5df0da](https://github.com/mljs/ml/commit/c5df0da))
110 | * **package:** update ml-kmeans to version 4.0.0 ([#101](https://github.com/mljs/ml/issues/101)) ([57c4a90](https://github.com/mljs/ml/commit/57c4a90))
111 | * **readme:** fix link to ml-random ([3a4b5c3](https://github.com/mljs/ml/commit/3a4b5c3))
112 | 
113 | 
114 | ### Features
115 | 
116 | * ML.Random ([6e44339](https://github.com/mljs/ml/commit/6e44339))
117 | 
118 | 
119 | 
120 | <a name="3.2.0"></a>
121 | # [3.2.0](https://github.com/mljs/ml/compare/v3.1.0...v3.2.0) (2018-02-21)
122 | 
123 | 
124 | ### Features
125 | 
126 | * add random forest regression and classification ([#94](https://github.com/mljs/ml/issues/94)) ([11688e3](https://github.com/mljs/ml/commit/11688e3))
127 | 
128 | 
129 | 
130 | <a name="3.1.0"></a>
131 | # [3.1.0](https://github.com/mljs/ml/compare/v3.0.0...v3.1.0) (2018-02-17)
132 | 
133 | 
134 | ### Features
135 | 
136 | * add ml-cart to the package ([#93](https://github.com/mljs/ml/issues/93)) ([001fc12](https://github.com/mljs/ml/commit/001fc12))
137 | 
138 | 
139 | 
140 | <a name="3.0.0"></a>
141 | # [3.0.0](https://github.com/mljs/ml/compare/v2.2.0...v3.0.0) (2018-02-17)
142 | 
143 | 
144 | ### Bug Fixes
145 | 
146 | * **package:** update ml-matrix to version 5.0.0 ([#76](https://github.com/mljs/ml/issues/76)) ([185d1c3](https://github.com/mljs/ml/commit/185d1c3))
147 | * **package:** update ml-naivebayes to version 3.0.0 ([3f312dc](https://github.com/mljs/ml/commit/3f312dc))
148 | * **package:** update ml-pls to version 1.0.0 ([7cf908a](https://github.com/mljs/ml/commit/7cf908a))
149 | 
150 | 
151 | 
152 | <a name="2.2.0"></a>
153 | # [2.2.0](https://github.com/mljs/ml/compare/v2.1.1...v2.2.0) (2017-07-14)
154 | 
155 | 
156 | ### Features
157 | 
158 | * add ml-confusion-matrix ([4af2e50](https://github.com/mljs/ml/commit/4af2e50))
159 | 
160 | 
161 | 
162 | <a name="2.1.1"></a>
163 | ## [2.1.1](https://github.com/mljs/ml/compare/v2.1.0...v2.1.1) (2017-06-29)
164 | 
165 | 
166 | 
167 | <a name="2.1.0"></a>
168 | # [2.1.0](https://github.com/mljs/ml/compare/v2.0.0...v2.1.0) (2017-06-23)
169 | 
170 | 
171 | 
172 | <a name="2.0.0"></a>
173 | # [2.0.0](https://github.com/mljs/ml/compare/v1.4.0...v2.0.0) (2017-02-10)
174 | 
175 | 
176 | ### Bug Fixes
177 | 
178 | * **package:** update ml-fnn to version 3.0.0 ([#59](https://github.com/mljs/ml/issues/59)) ([e7dc062](https://github.com/mljs/ml/commit/e7dc062))
179 | 
180 | 
181 | 
182 | <a name="1.4.0"></a>
183 | # [1.4.0](https://github.com/mljs/ml/compare/v1.3.1...v1.4.0) (2016-09-23)
184 | 
185 | 
186 | 
187 | <a name="1.3.1"></a>
188 | ## [1.3.1](https://github.com/mljs/ml/compare/v1.3.0...v1.3.1) (2016-09-05)
189 | 
190 | 
191 | 
192 | <a name="1.3.0"></a>
193 | # [1.3.0](https://github.com/mljs/ml/compare/v1.2.1...v1.3.0) (2016-09-05)
194 | 
195 | 
196 | ### Features
197 | 
198 | * **binary-search:** add v1.1.0 of binary search algorithm ([2a6d349](https://github.com/mljs/ml/commit/2a6d349))
199 | 
200 | 
201 | 
202 | <a name="1.2.1"></a>
203 | ## [1.2.1](https://github.com/mljs/ml/compare/v1.2.0...v1.2.1) (2016-08-16)
204 | 
205 | 
206 | 
207 | <a name="1.2.0"></a>
208 | # [1.2.0](https://github.com/mljs/ml/compare/v1.1.1...v1.2.0) (2016-08-11)
209 | 
210 | 
211 | ### Bug Fixes
212 | 
213 | * **ml-stat:** new access route for array and matrix ([3094e35](https://github.com/mljs/ml/commit/3094e35))
214 | 
215 | 
216 | ### Features
217 | 
218 | * add optimization package ([ceb5319](https://github.com/mljs/ml/commit/ceb5319))
219 | 
220 | 
221 | 
222 | <a name="1.1.1"></a>
223 | ## [1.1.1](https://github.com/mljs/ml/compare/v1.1.0...v1.1.1) (2016-08-03)
224 | 
225 | 
226 | 
227 | <a name="1.1.0"></a>
228 | # [1.1.0](https://github.com/mljs/ml/compare/v1.0.0...v1.1.0) (2016-08-03)
229 | 
230 | 
231 | 
232 | <a name="1.0.0"></a>
233 | # [1.0.0](https://github.com/mljs/ml/compare/v0.13.7...v1.0.0) (2016-06-22)
234 | 
235 | 
236 | 
237 | <a name="0.13.7"></a>
238 | ## [0.13.7](https://github.com/mljs/ml/compare/v0.13.6...v0.13.7) (2016-06-16)
239 | 
240 | 
241 | 
242 | <a name="0.13.6"></a>
243 | ## [0.13.6](https://github.com/mljs/ml/compare/v0.13.4...v0.13.6) (2016-06-15)
244 | 
245 | 
246 | 
247 | <a name="0.13.4"></a>
248 | ## [0.13.4](https://github.com/mljs/ml/compare/v0.13.3...v0.13.4) (2016-06-13)
249 | 
250 | 
251 | 
252 | <a name="0.13.3"></a>
253 | ## [0.13.3](https://github.com/mljs/ml/compare/v0.13.2...v0.13.3) (2016-06-13)
254 | 
255 | 
256 | 
257 | <a name="0.13.2"></a>
258 | ## [0.13.2](https://github.com/mljs/ml/compare/v0.13.1...v0.13.2) (2016-06-10)
259 | 
260 | 
261 | 
262 | <a name="0.13.0"></a>
263 | # [0.13.0](https://github.com/mljs/ml/compare/v0.12.0...v0.13.0) (2016-06-10)
264 | 
265 | 
266 | 
267 | <a name="0.12.0"></a>
268 | # [0.12.0](https://github.com/mljs/ml/compare/v0.11.2...v0.12.0) (2016-06-09)
269 | 
270 | 
271 | 
272 | <a name="0.11.2"></a>
273 | ## [0.11.2](https://github.com/mljs/ml/compare/v0.11.1...v0.11.2) (2016-06-02)
274 | 
275 | 
276 | 
277 | <a name="0.11.1"></a>
278 | ## [0.11.1](https://github.com/mljs/ml/compare/v0.11.0...v0.11.1) (2016-05-31)
279 | 
280 | 
281 | 
282 | <a name="0.11.0"></a>
283 | # [0.11.0](https://github.com/mljs/ml/compare/v0.10.1...v0.11.0) (2016-05-31)
284 | 
285 | 
286 | 
287 | <a name="0.10.1"></a>
288 | ## [0.10.1](https://github.com/mljs/ml/compare/v0.10.0...v0.10.1) (2016-05-30)
289 | 
290 | 
291 | 
292 | <a name="0.10.0"></a>
293 | # [0.10.0](https://github.com/mljs/ml/compare/v0.9.1...v0.10.0) (2016-05-26)
294 | 
295 | 
296 | 
297 | <a name="0.9.1"></a>
298 | ## [0.9.1](https://github.com/mljs/ml/compare/v0.9.0...v0.9.1) (2016-05-23)
299 | 
300 | 
301 | 
302 | <a name="0.9.0"></a>
303 | # [0.9.0](https://github.com/mljs/ml/compare/v0.8.2...v0.9.0) (2016-05-19)
304 | 
305 | 
306 | 
307 | <a name="0.8.2"></a>
308 | ## [0.8.2](https://github.com/mljs/ml/compare/v0.8.1...v0.8.2) (2016-05-14)
309 | 
310 | 
311 | 
312 | <a name="0.8.1"></a>
313 | ## [0.8.1](https://github.com/mljs/ml/compare/v0.8.0...v0.8.1) (2015-11-19)
314 | 
315 | 
316 | 
317 | <a name="0.8.0"></a>
318 | # [0.8.0](https://github.com/mljs/ml/compare/v0.7.0...v0.8.0) (2015-09-15)
319 | 
320 | 
321 | 
322 | <a name="0.7.0"></a>
323 | # [0.7.0](https://github.com/mljs/ml/compare/v0.6.2...v0.7.0) (2015-09-07)
324 | 
325 | 
326 | 
327 | <a name="0.6.2"></a>
328 | ## [0.6.2](https://github.com/mljs/ml/compare/v0.6.1...v0.6.2) (2015-09-02)
329 | 
330 | 
331 | 
332 | <a name="0.6.1"></a>
333 | ## [0.6.1](https://github.com/mljs/ml/compare/v0.6.0...v0.6.1) (2015-09-01)
334 | 
335 | 
336 | 
337 | <a name="0.6.0"></a>
338 | # [0.6.0](https://github.com/mljs/ml/compare/v0.5.1...v0.6.0) (2015-08-20)
339 | 
340 | 
341 | 
342 | <a name="0.5.1"></a>
343 | ## [0.5.1](https://github.com/mljs/ml/compare/v0.5.0...v0.5.1) (2015-08-20)
344 | 
345 | 
346 | 
347 | <a name="0.5.0"></a>
348 | # [0.5.0](https://github.com/mljs/ml/compare/v0.4.1...v0.5.0) (2015-08-20)
349 | 
350 | 
351 | 
352 | <a name="0.4.1"></a>
353 | ## [0.4.1](https://github.com/mljs/ml/compare/v0.4.0...v0.4.1) (2015-08-05)
354 | 
355 | 
356 | 
357 | <a name="0.4.0"></a>
358 | # [0.4.0](https://github.com/mljs/ml/compare/v0.3.10...v0.4.0) (2015-07-24)
359 | 
360 | 
361 | 
362 | <a name="0.3.10"></a>
363 | ## [0.3.10](https://github.com/mljs/ml/compare/v0.3.9...v0.3.10) (2015-07-08)
364 | 
365 | 
366 | 
367 | <a name="0.3.8"></a>
368 | ## [0.3.8](https://github.com/mljs/ml/compare/v0.3.7...v0.3.8) (2015-07-07)
369 | 
370 | 
371 | 
372 | <a name="0.3.7"></a>
373 | ## [0.3.7](https://github.com/mljs/ml/compare/v0.3.6...v0.3.7) (2015-07-03)
374 | 
375 | 
376 | 
377 | <a name="0.3.6"></a>
378 | ## [0.3.6](https://github.com/mljs/ml/compare/v0.3.5...v0.3.6) (2015-07-01)
379 | 
380 | 
381 | 
382 | <a name="0.3.5"></a>
383 | ## [0.3.5](https://github.com/mljs/ml/compare/v0.3.4...v0.3.5) (2015-06-30)
384 | 
385 | 
386 | 
387 | <a name="0.3.4"></a>
388 | ## [0.3.4](https://github.com/mljs/ml/compare/v0.3.3...v0.3.4) (2015-06-29)
389 | 
390 | 
391 | 
392 | <a name="0.3.3"></a>
393 | ## [0.3.3](https://github.com/mljs/ml/compare/v0.3.1...v0.3.3) (2015-06-23)
394 | 
395 | 
396 | 
397 | <a name="0.3.1"></a>
398 | ## [0.3.1](https://github.com/mljs/ml/compare/v0.3.0...v0.3.1) (2015-06-17)
399 | 
400 | 
401 | 
402 | <a name="0.3.0"></a>
403 | # [0.3.0](https://github.com/mljs/ml/compare/v0.2.3...v0.3.0) (2015-06-17)
404 | 
405 | 
406 | 
407 | <a name="0.2.3"></a>
408 | ## [0.2.3](https://github.com/mljs/ml/compare/v0.2.2...v0.2.3) (2015-04-24)
409 | 
410 | 
411 | 
412 | <a name="0.2.2"></a>
413 | ## [0.2.2](https://github.com/mljs/ml/compare/v0.2.1...v0.2.2) (2015-03-16)
414 | 
415 | 
416 | 
417 | <a name="0.2.1"></a>
418 | ## [0.2.1](https://github.com/mljs/ml/compare/v0.2.0...v0.2.1) (2015-02-02)
419 | 
420 | 
421 | 
422 | <a name="0.2.0"></a>
423 | # [0.2.0](https://github.com/mljs/ml/compare/v0.1.4...v0.2.0) (2015-01-27)
424 | 
425 | 
426 | 
427 | <a name="0.1.4"></a>
428 | ## [0.1.4](https://github.com/mljs/ml/compare/v0.1.3...v0.1.4) (2014-12-09)
429 | 
430 | 
431 | 
432 | <a name="0.1.3"></a>
433 | ## [0.1.3](https://github.com/mljs/ml/compare/v0.1.2...v0.1.3) (2014-12-09)
434 | 
435 | 
436 | 
437 | <a name="0.1.2"></a>
438 | ## [0.1.2](https://github.com/mljs/ml/compare/v0.1.1...v0.1.2) (2014-12-03)
439 | 
440 | 
441 | 
442 | <a name="0.1.1"></a>
443 | ## [0.1.1](https://github.com/mljs/ml/compare/v0.1.0...v0.1.1) (2014-12-01)
444 | 
445 | 
446 | 
447 | <a name="0.1.0"></a>
448 | # [0.1.0](https://github.com/mljs/ml/compare/v0.0.5...v0.1.0) (2014-11-18)
449 | 
450 | 
451 | 
452 | <a name="0.0.5"></a>
453 | ## [0.0.5](https://github.com/mljs/ml/compare/v0.0.4...v0.0.5) (2014-10-28)
454 | 
455 | 
456 | 
457 | <a name="0.0.4"></a>
458 | ## [0.0.4](https://github.com/mljs/ml/compare/v0.0.3...v0.0.4) (2014-10-28)
459 | 
460 | 
461 | 
462 | <a name="0.0.3"></a>
463 | ## [0.0.3](https://github.com/mljs/ml/compare/v0.0.2...v0.0.3) (2014-10-28)
464 | 
465 | 
466 | 
467 | <a name="0.0.2"></a>
468 | ## [0.0.2](https://github.com/mljs/ml/compare/v0.0.1...v0.0.2) (2014-10-28)
469 | 
470 | 
471 | 
472 | <a name="0.0.1"></a>
473 | ## 0.0.1 (2014-10-28)
474 | 


--------------------------------------------------------------------------------
/CODE_OF_CONDUCT.md:
--------------------------------------------------------------------------------
 1 | # Contributor Covenant Code of Conduct
 2 | 
 3 | ## Our Pledge
 4 | 
 5 | In the interest of fostering an open and welcoming environment, we as contributors and maintainers pledge to making participation in our project and our community a harassment-free experience for everyone, regardless of age, body size, disability, ethnicity, gender identity and expression, level of experience, nationality, personal appearance, race, religion, or sexual identity and orientation.
 6 | 
 7 | ## Our Standards
 8 | 
 9 | Examples of behavior that contributes to creating a positive environment include:
10 | 
11 | * Using welcoming and inclusive language
12 | * Being respectful of differing viewpoints and experiences
13 | * Gracefully accepting constructive criticism
14 | * Focusing on what is best for the community
15 | * Showing empathy towards other community members
16 | 
17 | Examples of unacceptable behavior by participants include:
18 | 
19 | * The use of sexualized language or imagery and unwelcome sexual attention or advances
20 | * Trolling, insulting/derogatory comments, and personal or political attacks
21 | * Public or private harassment
22 | * Publishing others' private information, such as a physical or electronic address, without explicit permission
23 | * Other conduct which could reasonably be considered inappropriate in a professional setting
24 | 
25 | ## Our Responsibilities
26 | 
27 | Project maintainers are responsible for clarifying the standards of acceptable behavior and are expected to take appropriate and fair corrective action in response to any instances of unacceptable behavior.
28 | 
29 | Project maintainers have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, or to ban temporarily or permanently any contributor for other behaviors that they deem inappropriate, threatening, offensive, or harmful.
30 | 
31 | ## Scope
32 | 
33 | This Code of Conduct applies both within project spaces and in public spaces when an individual is representing the project or its community. Examples of representing a project or community include using an official project e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event. Representation of a project may be further defined and clarified by project maintainers.
34 | 
35 | ## Enforcement
36 | 
37 | Instances of abusive, harassing, or otherwise unacceptable behavior may be reported by contacting the project team at admin@cheminfo.org. The project team will review and investigate all complaints, and will respond in a way that it deems appropriate to the circumstances. The project team is obligated to maintain confidentiality with regard to the reporter of an incident. Further details of specific enforcement policies may be posted separately.
38 | 
39 | Project maintainers who do not follow or enforce the Code of Conduct in good faith may face temporary or permanent repercussions as determined by other members of the project's leadership.
40 | 
41 | ## Attribution
42 | 
43 | This Code of Conduct is adapted from the [Contributor Covenant][homepage], version 1.4, available at [http://contributor-covenant.org/version/1/4][version]
44 | 
45 | [homepage]: http://contributor-covenant.org
46 | [version]: http://contributor-covenant.org/version/1/4/
47 | 


--------------------------------------------------------------------------------
/CONTRIBUTING.md:
--------------------------------------------------------------------------------
  1 | # Contributing to mljs
  2 | 
  3 | This document explains the general guidelines for the development of JavaScript APIs
  4 | for the machine learning algorithms.
  5 | 
  6 |   * [General rules](#general-rules)
  7 |   * [Standalone functions](#standalone-functions)
  8 |   * [Predictors](#predictors)
  9 |   * [Commit Guidelines](#commit-guidelines)
 10 |   * [Issues and Bugs](#issues-and-bugs)
 11 | 
 12 | ## General rules
 13 | 
 14 | ### Use camelCase
 15 | 
 16 | For consistency and because it is by far the most common style in JavaScript.  
 17 | Exception: class names must start with a capital letter.
 18 | 
 19 | #### Good
 20 | 
 21 | * xSquared
 22 | * kernelType
 23 | * maybeToPrecision
 24 | 
 25 | #### Bad
 26 | 
 27 | * x_squared
 28 | * KernelType
 29 | * maybe_To$precision
 30 | 
 31 | ### Use understandable names
 32 | 
 33 | The API is the point of contact with the outside world. People who use it should understand what an
 34 | option or a function does without looking at the documentation.
 35 | 
 36 | #### Good
 37 | 
 38 | * alpha, beta, gamma, ...
 39 | * kernel, xSquared, numerator
 40 | 
 41 | #### Bad
 42 | 
 43 | * a, b, c
 44 | * k, x2, num
 45 | 
 46 | ## Standalone functions
 47 | 
 48 | Functions that take some input and directly return the result should always have the following signature:
 49 | 
 50 | ```js
 51 | function myFunction(param1, param2, ..., paramN, options = {}) { ... }
 52 | ```
 53 | 
 54 | The `param1` to `paramN` arguments are reserved for __mandatory__ parameters. Anything else goes in an `options` object.  
 55 | The call should not fail if `options` is undefined.
 56 | 
 57 | To handle default options, use default parameters and object destructuring:
 58 | 
 59 | ```js
 60 | function myFunction(options = {}) {
 61 |     const {
 62 |         option1 = 'value1',
 63 |         option2 = 'value2'
 64 |     } = options;
 65 |     ...
 66 | }
 67 | ```
 68 | 
 69 | ## Predictors
 70 | 
 71 | Predictors are classes which implement the following interface:
 72 | 
 73 | ### new Predictor([options])
 74 | 
 75 | Creates the predictor. The constructor can take parameters or options to initialize the algorithm.  
 76 | Alternatively, if the predictor has no training phase, it can be instantiated like so: `new Predictor(features[[, labels], options])`
 77 | 
 78 | ### predictor.train(features[[, labels], options])
 79 | If the predictor has a training phase, it is executed here.
 80 | 
 81 | ### predictor.predict(features)
 82 | 
 83 | This method runs the prediction for a new set of observations.
 84 | 
 85 | ### predictor.score()
 86 | 
 87 | This method is optional.
 88 | It should return a value that represents the quality of a predictor.
 89 | 
 90 | ### predictor.toJSON()
 91 | 
 92 | This method should return plain JavaScript Object that enables to reload the current predictor
 93 | and that can be serialized to a JSON string using `JSON.stringify`
 94 | 
 95 | ### Predictor.load(json)
 96 | 
 97 | This static method should return a new predictor instance that is ready to make predictions. The `json`
 98 | parameter is the object returned by an earlier call of `toJSON`.
 99 | 
100 | ## Commit Guidelines
101 | 
102 | The rules are based on the [AngularJS commit guidelines](https://github.com/angular/angular.js/blob/main/CONTRIBUTING.md#commit). This leads to **more readable messages** that are easy to follow when looking through the **project history**.
103 | 
104 | ### Commit Message Format
105 | Each commit message consists of a **header**, a **body** and a **footer**.  The header has a special
106 | format that includes a **type**, a **scope** and a **subject**:
107 | 
108 | ```
109 | <type>(<scope>): <subject>
110 | <BLANK LINE>
111 | <body>
112 | <BLANK LINE>
113 | <footer>
114 | ```
115 | 
116 | The **header** is mandatory and the **scope** of the header is optional.
117 | 
118 | Any line of the commit message cannot be longer 100 characters! This allows the message to be easier
119 | to read on GitHub as well as in various git tools.
120 | 
121 | ### Revert
122 | If the commit reverts a previous commit, it should begin with `revert: `, followed by the header of the reverted commit. In the body it should say: `This reverts commit <hash>.`, where the hash is the SHA of the commit being reverted.
123 | 
124 | ### Type
125 | Must be one of the following:
126 | 
127 | * **feat**: A new feature
128 | * **fix**: A bug fix
129 | * **docs**: Documentation only changes
130 | * **style**: Changes that do not affect the meaning of the code (white-space, formatting, missing
131 |   semi-colons, etc)
132 | * **refactor**: A code change that neither fixes a bug nor adds a feature
133 | * **perf**: A code change that improves performance
134 | * **test**: Adding missing tests
135 | * **chore**: Changes to the build process or auxiliary tools and libraries such as documentation
136 |   generation
137 | 
138 | ### Scope
139 | The scope could be anything specifying place of the commit change.
140 | 
141 | ### Subject
142 | The subject contains succinct description of the change:
143 | 
144 | * use the imperative, present tense: "change" not "changed" nor "changes"
145 | * don't capitalize first letter
146 | * no dot (.) at the end
147 | 
148 | ### Body
149 | Just as in the **subject**, use the imperative, present tense: "change" not "changed" nor "changes".
150 | The body should include the motivation for the change and contrast this with previous behavior.
151 | 
152 | ### Footer
153 | The footer should contain any information about **Breaking Changes** and is also the place to
154 | reference GitHub issues that this commit **Closes**.
155 | 
156 | **Breaking Changes** should start with the word `BREAKING CHANGE:` with a space or two newlines. The rest of the commit message is then used for this.
157 | 
158 | ## Issues and Bugs
159 | 
160 | This repository is a wrap of several modules, so if the issue is related to an specific module,
161 | please create the issue in that repository. This will make easier and faster the solution of that.
162 |  
163 | Before you submit your issue search the archive, maybe your question was already answered.
164 | 
165 | If your issue appears to be a bug, and hasn't been reported, open a new issue.
166 | Help us to maximize the effort we can spend fixing issues and adding new
167 | features, by not reporting duplicate issues.  Providing the following information will increase the
168 | chances of your issue being dealt with quickly:
169 | 
170 | * **Overview of the Issue** - if an error is being thrown a non-minified stack trace helps
171 | * **Motivation for or Use Case** - explain why this is a bug for you
172 | * **Module Version(s)** - is it a regression?
173 | * **NodeJS Version(s) and Operating System** - is this a platform related issue?
174 | * **Reproduce the Error** - provide a live example (using [Tonic](https://tonicdev.com/) or
175 |   [Visualizer](https://my.cheminfo.org)) or an unambiguous set of steps.
176 | * **Related Issues** - has a similar issue been reported before?
177 | * **Suggest a Fix** - if you can't fix the bug yourself, perhaps you can point to what might be
178 |   causing the problem (line of code or commit)
179 | 


--------------------------------------------------------------------------------
/LICENSE:
--------------------------------------------------------------------------------
 1 | The MIT License (MIT)
 2 | 
 3 | Copyright (c) 2015 ml.js
 4 | 
 5 | Permission is hereby granted, free of charge, to any person obtaining a copy
 6 | of this software and associated documentation files (the "Software"), to deal
 7 | in the Software without restriction, including without limitation the rights
 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 9 | copies of the Software, and to permit persons to whom the Software is
10 | furnished to do so, subject to the following conditions:
11 | 
12 | The above copyright notice and this permission notice shall be included in all
13 | copies or substantial portions of the Software.
14 | 
15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21 | SOFTWARE.
22 | 
23 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
  1 | # ml.js - Machine learning tools in JavaScript
  2 | 
  3 | ## Introduction
  4 | 
  5 | This library is a compilation of the tools developed in the [mljs](https://github.com/mljs) organization.  
  6 | It is mainly maintained for use in the browser. If you are working with Node.js, you might prefer to add
  7 | to your dependencies only the libraries that you need, as they are usually published to npm more often.  
  8 | We prefix all our npm package names with `ml-` (eg. ml-matrix) so they are easy to find.
  9 | 
 10 | To include the ml.js library in a web page:
 11 | 
 12 | ```html
 13 | <script src="https://www.lactame.com/lib/ml/6.0.0/ml.min.js"></script>
 14 | ```
 15 | 
 16 | It will be available as the global `ML` variable. The package is in UMD format.
 17 | 
 18 | ## List of included libraries
 19 | 
 20 | ### Unsupervised learning
 21 | 
 22 | - Principal component analysis (PCA): [`ML.PCA`](https://github.com/mljs/pca)
 23 | - Hierarchical clustering: [`ML.HClust`](https://github.com/mljs/hclust)
 24 | - K-means clustering: [`ML.KMeans`](https://github.com/mljs/kmeans)
 25 | 
 26 | ### Supervised learning
 27 | 
 28 | - Naive Bayes: [`ML.NaiveBayes`](https://github.com/mljs/naive-bayes)
 29 | - K-Nearest Neighbor (KNN): [`ML.KNN`](https://github.com/mljs/knn)
 30 | - Partial least squares (PLS): [`ML.PLS`](https://github.com/mljs/pls)
 31 | - K-OPLS: [`ML.KOPLS`](https://github.com/mljs/pls)
 32 | - Cross-validation: [`ML.CrossValidation`](https://github.com/mljs/cross-validation)
 33 | - Confusion matrix: [`ML.ConfusionMatrix`](https://github.com/mljs/confusion-matrix)
 34 | - Decision tree classifier: [`ML.DecisionTreeClassifier`](https://github.com/mljs/decision-tree-cart)
 35 | - Random forest classifier: [`ML.RandomForestClassifier`](https://github.com/mljs/random-forest)
 36 | 
 37 | ### Artificial neural networks (ANN)
 38 | 
 39 | - Feedforward Neural Networks: [`ML.FNN`](https://github.com/mljs/feedforward-neural-networks)
 40 | - Self-organizing map / Kohonen networks: [`ML.SOM`](https://github.com/mljs/som)
 41 | 
 42 | ### Regression
 43 | 
 44 | - Simple linear regression: [`ML.SimpleLinearRegression`](https://github.com/mljs/regression-simple-linear)
 45 | - Polynomial regression: [`ML.PolynomialRegression`](https://github.com/mljs/regression-polynomial)
 46 | - Multivariate linear regression: [`ML.MultivariateLinearRegression`](https://github.com/mljs/regression-multivariate-linear)
 47 | - Power regression: [`ML.PowerRegression`](https://github.com/mljs/regression-power)
 48 | - Exponential regression: [`ML.ExponentialRegression`](https://github.com/mljs/regression-exponential)
 49 | - Theil-Sen regression: [`ML.TheilSenRegression`](https://github.com/mljs/regression-theil-sen)
 50 | - Robust polynomial regression: [`ML.RobustPolynomialRegression`](https://github.com/mljs/regression-robust-polynomial)
 51 | - Decision tree regression: [`ML.DecisionTreeRegression`](https://github.com/mljs/decision-tree-cart)
 52 | - Random forest regression: [`ML.RandomForestRegression`](https://github.com/mljs/random-forest)
 53 | 
 54 | ### Optimization
 55 | 
 56 | - Levenberg-Marquardt: [`ML.levenbergMarquardt`](https://github.com/mljs/levenberg-marquardt)
 57 | - Fast Combinatorial Non-negative Least Squares: [`ML.FCNNLS`](https://github.com/mljs/fcnnls)
 58 | 
 59 | ### Math
 60 | 
 61 | - Matrix: [`ML.Matrix`](https://github.com/mljs/matrix) (Matrix class)
 62 | - Singular value decomposition (SVD): [`ML.SVD`](https://github.com/mljs/matrix)
 63 | - Eigenvalue decomposition (EVD): [`ML.EVD`](https://github.com/mljs/matrix)
 64 | - Cholesky decomposition: [`ML.CholeskyDecomposition`](https://github.com/mljs/matrix)
 65 | - Lu decomposition: [`ML.LuDecomposition`](https://github.com/mljs/matrix)
 66 | - Qr decomposition: [`ML.QrDecomposition`](https://github.com/mljs/matrix)
 67 | - Sparse matrix: [`ML.SparseMatrix`](https://github.com/mljs/sparse-matrix)
 68 | - Kernels: [`ML.Kernel`](https://github.com/mljs/kernel)
 69 | - Distance functions: [`ML.Distance`](https://github.com/mljs/distance)
 70 | - Similarity functions: [`ML.Similarity`](https://github.com/mljs/distance)
 71 | - Distance matrix: [`ML.distanceMatrix`](https://github.com/mljs/distance-matrix)
 72 | - XORShift-add RNG: [`ML.XSadd`](https://github.com/mljs/xsadd)
 73 | - non-negative Generalized Morphological Component Analysis [`ML.nGMCA`](https://github.com/mljs/nGMCA)
 74 | 
 75 | ### [`ML.Array`](https://github.com/mljs/array)
 76 | 
 77 | - [`ML.Array.min`](https://github.com/mljs/array/tree/master/packages/array-min)
 78 | - [`ML.Array.max`](https://github.com/mljs/array/tree/master/packages/array-max)
 79 | - [`ML.Array.median`](https://github.com/mljs/array/tree/master/packages/array-median)
 80 | - [`ML.Array.mean`](https://github.com/mljs/array/tree/master/packages/array-mean)
 81 | - [`ML.Array.mode`](https://github.com/mljs/array/tree/master/packages/array-mode)
 82 | - [`ML.Array.normed`](https://github.com/mljs/array/tree/master/packages/array-normed)
 83 | - [`ML.Array.rescale`](https://github.com/mljs/array/tree/master/packages/array-rescale)
 84 | - [`ML.Array.sequentialFill`](https://github.com/mljs/array/tree/master/packages/array-sequential-fill)
 85 | - [`ML.Array.standardDeviation`](https://github.com/mljs/array/tree/master/packages/array-standard-deviation)
 86 | - [`ML.Array.variance`](https://github.com/mljs/array/tree/master/packages/array-variance)
 87 | 
 88 | ### [`ML.ArrayXY`](https://github.com/mljs/array-xy)
 89 | 
 90 | Functions dealing with an object containing 2 properties x and y, both arrays.
 91 | 
 92 | Example:
 93 | 
 94 | ```js
 95 | let result = ML.ArrayXY.sortX({ x: [2, 3, 1], y: [4, 6, 2] });
 96 | // result = {x: [1,2,3], y: [2,4,6]}
 97 | ```
 98 | 
 99 | - [ML.ArrayXY.weightedMerge](https://github.com/mljs/array-xy/tree/master/packages/array-xy-weighted-merge): Merge abscissa values on similar ordinates and weight the group of abscissa
100 | 
101 | - [ML.ArrayXY.maxMerge](https://github.com/mljs/array-xy/tree/master/packages/array-xy-max-merge): Merge abscissa values on similar ordinates and keeps the abscissa with bigger ordinate value
102 | - [ML.ArrayXY.closestX](https://github.com/mljs/array-xy/tree/master/packages/array-xy-closest-x): Get the closest point for a specific abscissa value
103 | - [ML.ArrayXY.centroidsMerge](https://github.com/mljs/array-xy/tree/master/packages/array-xy-centroids-merge): Merge abscissa values if the ordinate value is in a list of centroids
104 | - [ML.ArrayXY.sortX](https://github.com/mljs/array-xy/tree/master/packages/array-xy-sort-x): Sort a set of point based on the abscissas values
105 | - [ML.ArrayXY.maxY](https://github.com/mljs/array-xy/tree/master/packages/array-xy-max-y): Sort a set of point based on the abscissas values
106 | - [ML.ArrayXY.uniqueX](https://github.com/mljs/array-xy/tree/master/packages/array-xy-unique-x): Ensure that x values are unique
107 | 
108 | ### Statistics
109 | 
110 | - Performance (ROC curve): [`ML.Performance`](https://github.com/mljs/performance)
111 | 
112 | ### Data processing
113 | 
114 | - Principal component analysis (PCA): [`ML.PCA`](https://github.com/mljs/pca)
115 | - Savitzky-Golay filter: [`ML.savitzkyGolay`](https://github.com/mljs/savitzky-golay)
116 | - Global Spectral Deconvolution (GSD): [`ML.GSD`](https://github.com/mljs/global-spectral-deconvolution)
117 | 
118 | ### Utility
119 | 
120 | - Bit array operations: [`ML.BitArray`](https://github.com/mljs/bit-array)
121 | - Hash table: [`ML.HashTable`](https://github.com/mljs/hash-table)
122 | - Pad array: [`ML.padArray`](https://github.com/mljs/pad-array)
123 | - Binary search: [`ML.binarySearch`](https://github.com/darkskyapp/binary-search)
124 | - Number comparison functions for sorting: [`ML.numSort`](https://github.com/sindresorhus/num-sort)
125 | - Random number generation: [`ML.Random`](https://github.com/mljs/random)
126 | 
127 | ## License
128 | 
129 | [MIT](./LICENSE)
130 | 


--------------------------------------------------------------------------------
/babel.config.json:
--------------------------------------------------------------------------------
1 | {
2 |   "plugins": ["@babel/plugin-transform-modules-commonjs"]
3 | }
4 | 


--------------------------------------------------------------------------------
/bower.json:
--------------------------------------------------------------------------------
 1 | {
 2 |   "name": "ml",
 3 |   "version": "5.3.0",
 4 |   "main": [
 5 |     "dist/ml.js",
 6 |     "dist/ml.min.js"
 7 |   ],
 8 |   "homepage": "https://github.com/mljs/ml",
 9 |   "authors": [
10 |     "Michaël Zasso"
11 |   ],
12 |   "description": "Machine learning tools",
13 |   "moduleType": [
14 |     "amd",
15 |     "globals",
16 |     "node"
17 |   ],
18 |   "keywords": [
19 |     "machine",
20 |     "learning",
21 |     "data",
22 |     "mining",
23 |     "datamining"
24 |   ],
25 |   "license": "MIT",
26 |   "ignore": [
27 |     "/*",
28 |     "!dist/*"
29 |   ]
30 | }
31 | 


--------------------------------------------------------------------------------
/eslint.config.mjs:
--------------------------------------------------------------------------------
 1 | import cheminfo from 'eslint-config-cheminfo';
 2 | import globals from 'globals';
 3 | 
 4 | export default [
 5 |   ...cheminfo,
 6 |   {
 7 |     languageOptions: {
 8 |       globals: {
 9 |         ...globals.node,
10 |       },
11 |     },
12 |     rules: {
13 |       //      "camelcase": "off",
14 |     }
15 |   }
16 | ]


--------------------------------------------------------------------------------
/examples/array.js:
--------------------------------------------------------------------------------
1 | const Array=require('../dist/ml.js').Array;
2 | 
3 | var min=Array.min([1,4,6,-1,2]);
4 | console.log(min);
5 | 
6 | 


--------------------------------------------------------------------------------
/examples/fcnnls.js:
--------------------------------------------------------------------------------
 1 | const { fcnnlsVector } = require('../dist/ml.js').FCNNLS;
 2 | 
 3 | // Example that allows to calculate the contribution of pure
 4 | // spectra in a mixture.
 5 | 
 6 | // columns correspond to pure sample
 7 | let matrix = [[1, 1, 1], [1, 1, 0], [1, 0, 0]];
 8 | 
 9 | let mixture = [5, 3, 1];
10 | 
11 | let result = fcnnlsVector(matrix, mixture);
12 | console.log(result);
13 | 
14 | // should be close to 1,2,2
15 | 


--------------------------------------------------------------------------------
/examples/leafDataset/README.md:
--------------------------------------------------------------------------------
  1 | Introduction to classification with mljs
  2 | ========================================
  3 | 
  4 | This tutorial will solve a problem of Machine Learning called classification. Classification is the problem of identifying to which of a set of categories a new observation belongs ([Wikipedia](https://en.wikipedia.org/wiki/Statistical_classification)).
  5 | 
  6 | We will use methods of supervised learning. Supervised learning is the machine learnin task of inferring a function fom labeled training data. What it means is that we will use data with labels to train our models, and next we will try to predict the labels of other data (called test data).
  7 | 
  8 | We will use functions from ML.JS (https://github.com/mljs). We will try 4 different methods : SVM, KNN, Logistic regression and naive bayes.
  9 | 
 10 | Dataset
 11 | =======
 12 | 
 13 | We have to work with data, so I've chosen a dataset among UCI Machine Learning repository. We will try to define the type of leafs thanks to their features.
 14 | 
 15 | Dataset used : https://archive.ics.uci.edu/ml/datasets/Leaf
 16 | 
 17 | 
 18 | In this dataset, we have 340 data, with 16 features each. The first feature is the class (i.e the type of the leaf, i.e the label). The goal of the classification is to determine the class of a given leaf thanks to the 15 others features of this leaf.
 19 | 
 20 | On the previous link, you can find the description of each feature. These features are float or number.
 21 | 
 22 | Note :
 23 | One of the difficulties of this dataset is that we haven't a lot of data.
 24 | 
 25 | 
 26 | Installation :
 27 | ==============
 28 | 
 29 | ```
 30 | npm install libsvm-js ml-naivebayes ml-knn ml-logistic-regression csvtojson 
 31 | ```
 32 | 
 33 | First step :
 34 | ============
 35 | 
 36 | The first step will be to load the dataset and to delete the useless feature. Indeed, the second feature is useless, because it's just created to identify a data. We will use the package csvtojson to import and manipulate the data.
 37 | 
 38 | The data are in a csv file without header. Here is how to import it :
 39 | 
 40 | ```javascript
 41 | const csv = require('csvtojson');
 42 | 
 43 | const csvFilePath = 'leaf.csv'; // Data
 44 | const names = ['type', 'specimenNumber', 'eccentricity', 'aspectRatio', 'elongation', 'solidity', 'stochasticConvexity', 'isoperimetricFactor', 'maxIndetationDepth', 'lobedness', 'intensity', 'contrast', 'smoothness', 'thirdMoment', 'uniformity', 'entropy']; // For header
 45 | 
 46 | csv({noheader: true, headers: names})
 47 |     .fromFile(csvFilePath)
 48 |     .on('json', (jsonObj) => {
 49 |         data.push(jsonObj); // Push each object to data Array
 50 |     })
 51 |     .on('done', (error) => {
 52 |         seperationSize = 0.9 * data.length;
 53 |         data = shuffleArray(data);
 54 |         dressData();
 55 |     });
 56 | 
 57 | function dressData() {
 58 |     let types = new Set(); // To gather UNIQUE classes
 59 |     data.forEach((row) => {
 60 |         types.add(row.type);
 61 |     });
 62 |     typesArray = [...types]; // To save the different types of classes.
 63 | 
 64 |     data.forEach((row) => {
 65 |         let rowArray, typeNumber;
 66 |         rowArray = Object.keys(row).map(key => parseFloat(row[key])).slice(2, 16); // We don't use the 2 first elements, which are the type (i.e class) and the specimen number (i.e ID)
 67 |         typeNumber = typesArray.indexOf(row.type); // Convert type(String) to type(Number)
 68 | 
 69 |         X.push(rowArray);
 70 |         y.push(typeNumber);
 71 |     });
 72 | 
 73 |     trainingSetX = X.slice(0, seperationSize);
 74 |     trainingSetY = y.slice(0, seperationSize);
 75 |     testSetX = X.slice(seperationSize);
 76 |     testSetY = y.slice(seperationSize);
 77 | 
 78 | }
 79 | ``` 
 80 | 
 81 | We store the whole dataset in the variable X (and the labels in the variable y), and then we have split them into training set and test set. At the end of this tutorial, I will show you a method called cross-validation which replaces this random splitting. Indeed, the splitting between training and test sets is random. That's why if you run many times the script, results will be different because the separation between training set and test set will be different.
 82 | 
 83 | Second step - Configuration of the model :
 84 | ==========================================
 85 | 
 86 | We will try four different methods : the SVM classifier, the KNN classifier, the logistic regression and the naive bayes classifier.
 87 | 
 88 | SVM
 89 | ---
 90 | 
 91 | Here is the source code of the instanciation of the model SVM :
 92 | 
 93 | ```javascript
 94 | const SVM = require('libsvm-js/asm');
 95 | 
 96 | let options = {
 97 |     kernel : SVM.KERNEL_TYPES.POLYNOMIAL,
 98 |     degree : 3
 99 | }
100 | 
101 | svm = new SVM(options);
102 | ```
103 | 
104 | The parameters of the SVM are very important. Results depends a lot of these parameters. There are three Kernels often used : The linear kernel, the polynomial kernel and the RBF kernel. The main parameter of the polynomial kernel is the degree, and the main parameters of the RBF kernel are gamma and the cost.
105 | 
106 | Bayes
107 | -----
108 | ```javascript
109 | const Bayes = require('ml-naivebayes');
110 | 
111 | bayes = new Bayes();
112 | ```
113 | 
114 | Note : We don't need to give parameters to the naive bayes classifier. It's easier to use than the others classifiers ! However, we can't use always this classifier.
115 | 
116 | KNN
117 | ---
118 | 
119 | Here is the line which instanciate the KNN :
120 | 
121 | ```javascript
122 | knn = new KNN(trainingSetX, trainingSetY, {k:5});
123 | ```
124 | 
125 | The implementation of the KNN is different from the others because we need to give the training set when we instanciate a KNN classifier. The parameter k is very important for the accuracy of the model.
126 | 
127 | 
128 | Logistic Regression
129 | -------------------
130 | 
131 | We need to use the type Matrix to use the logistic regression (instead of the Array 1D or 2D). That's why we need the package ml-matrix.
132 | ```javascript
133 | const LogisticRegression = require('ml-logistic-regression');
134 | const {Matrix} = require('ml-matrix');
135 | 
136 | let logreg = new LogisticRegression({numSteps: 10000, learningRate: 5e-3});
137 | ```
138 | 
139 | The parameters of the logistic regression are important too. The higher the number of steps is, the slower the training will be.
140 | 
141 | Training and Evaluation :
142 | =========================
143 | 
144 | We have our model. Now we can train it with the training data. I just write below the source code to train the models :
145 | 
146 | For the SVM :
147 | ```javascript
148 | svm.train(trainingSetX, trainingSetY);
149 | ```
150 | 
151 | For the naive bayes :
152 | ```javascript
153 | bayes.train(trainingSetX, trainingSetY);
154 | ```
155 | 
156 | For the KNN, the model is trained after the instanciation of the KNN-classifier, so we don't have the method *train*.
157 | 
158 | For the logistic regression, we use the matrix (I add the code which transforms the arrays into the type Matrix):
159 | ```javascript
160 | trainingSetX = new Matrix(X.slice(0, seperationSize));
161 | trainingSetY = Matrix.columnVector(y.slice(0, seperationSize));
162 | testSetX = new Matrix(X.slice(seperationSize));
163 | logreg.train(trainingSetX, trainingSetY);
164 | ```
165 | 
166 | Evaluation
167 | -----------
168 | 
169 | When the model is trained, we can use it on the test set. We will predict the labels of these data and compare the predicted labels with the expected labels. Here is how we do that :
170 | 
171 | ```javascript
172 | function test() {
173 |     const result = svm.predict(testSetX);
174 |     const testSetLength = testSetX.length
175 |     const predictionError = error(result, testSetY);
176 |     console.log(`Test Set Size = ${testSetLength} and number of Misclassifications = ${predictionError}`);
177 | }
178 | 
179 | function error(predicted, expected) {
180 |     let misclassifications = 0;
181 |     for (var index = 0; index < predicted.length; index++) {
182 |         console.log(`truth : ${expected[index]} and prediction : ${predicted[index]}`);
183 |         if (predicted[index] !== expected[index]) {
184 |             misclassifications++;
185 |         }
186 |     }
187 |     return misclassifications;
188 | }
189 | ```
190 | We first predict the labels thanks to the model, and then we use the function error() to compare the predicted labels with the real labels (so the expected labels).
191 | 
192 | The script will display for each data from the test set the expected labels and the predicted labels.
193 | 
194 | The accuracy is the percentage of right predictions. We have the number of test data (let's call it N) and the number of misclassification (let's call it f), so the accucary is (f/N)*100 %.
195 | 
196 | Results :
197 | =========
198 |  
199 | SVM
200 | ---
201 | There are on average 7 or 8 errors on 34 predictions, so the accuracy is ~78%. (the result oscillate between 4 and 14, it depends of the splitting into the training and test sets)
202 | 
203 | Bayes
204 | -----
205 | The results are on average between 8 and 10 errors on 34 predictions, so accuracy between 70% and 76%. 
206 | 
207 | KNN 
208 | ---
209 | With KNN (k=1), the result are on average between 11 and 15 errors, so accuracy between 60% and 70%.
210 | 
211 | Logistic Regression
212 | -------------------
213 | With 5000 steps and learning rate of 5e-3, we have between 9 and 11 errors on average, so accuracy between 68% and 75%. 
214 | 
215 | Conclusion :
216 | ============
217 | 
218 | This tutorial show an example of solving of a classification problem with SVM and naive bayes classifier. You can use a lot of others Machine Learning methods to solve this problem (random forests, neural networks...) and the parameters given to this SVM in my example are probably not the best, but the goal of this tutorial is only to let you see how to use ML.JS to solve classification problems (you can have better precision than 78%). The best accuracy with SVM will probably be with RBF, if you have a good couple of gamme/cost.
219 | 
220 | Source code :
221 | =============
222 | 
223 | SVM
224 | ----
225 | ```javascript
226 | const SVM = require('libsvm-js/asm');
227 | const csv = require('csvtojson');
228 | 
229 | let options = {
230 |     kernel : SVM.KERNEL_TYPES.POLYNOMIAL,
231 |     degree : 3
232 | }
233 | 
234 | const csvFilePath = 'leaf.csv'; // Data
235 | const names = ['type', 'specimenNumber', 'eccentricity', 'aspectRatio', 'elongation', 'solidity', 'stochasticConvexity', 'isoperimetricFactor', 'maxIndetationDepth', 'lobedness', 'intensity', 'contrast', 'smoothness', 'thirdMoment', 'uniformity', 'entropy']; // For header
236 | 
237 | svm = new SVM(options);
238 | 
239 | let seperationSize;
240 | let data = [], X = [], y = [];
241 | let trainingSetX = [], trainingSetY = [], testSetX = [], testSetY = [];
242 | 
243 | csv({noheader: true, headers: names})
244 |     .fromFile(csvFilePath)
245 |     .on('json', (jsonObj) => {
246 |         data.push(jsonObj); // Push each object to data Array
247 |     })
248 |     .on('done', (error) => {
249 |         seperationSize = 0.9 * data.length;
250 |         data = shuffleArray(data);
251 |         dressData();
252 |     });
253 | 
254 | function dressData() {
255 |     let types = new Set(); // To gather UNIQUE classes
256 |     data.forEach((row) => {
257 |         types.add(row.type);
258 |     });
259 |     typesArray = [...types]; // To save the different types of classes.
260 | 
261 |     data.forEach((row) => {
262 |         let rowArray, typeNumber;
263 |         rowArray = Object.keys(row).map(key => parseFloat(row[key])).slice(2, 16); // We don't use the 2 first elements, which are the type (i.e class) and the specimen number (i.e ID)
264 |         typeNumber = typesArray.indexOf(row.type); // Convert type(String) to type(Number)
265 | 
266 |         X.push(rowArray);
267 |         y.push(typeNumber);
268 |     });
269 | 
270 |     trainingSetX = X.slice(0, seperationSize);
271 |     trainingSetY = y.slice(0, seperationSize);
272 |     testSetX = X.slice(seperationSize);
273 |     testSetY = y.slice(seperationSize);
274 | 
275 |     train();
276 | }
277 | 
278 | function train() {
279 |     svm.train(trainingSetX, trainingSetY);
280 |     test();
281 | }
282 | 
283 | function test() {
284 |     const result = svm.predict(testSetX);
285 |     const testSetLength = testSetX.length
286 |     const predictionError = error(result, testSetY);
287 |     console.log(`Test Set Size = ${testSetLength} and number of Misclassifications = ${predictionError}`);
288 | }
289 | 
290 | function error(predicted, expected) {
291 |     let misclassifications = 0;
292 |     for (var index = 0; index < predicted.length; index++) {
293 |         console.log(`truth : ${expected[index]} and prediction : ${predicted[index]}`);
294 |         if (predicted[index] !== expected[index]) {
295 |             misclassifications++;
296 |         }
297 |     }
298 |     return misclassifications;
299 | }
300 | 
301 | 
302 | function shuffleArray(array) {
303 |     for (var i = array.length - 1; i > 0; i--) {
304 |         var j = Math.floor(Math.random() * (i + 1));
305 |         var temp = array[i];
306 |         array[i] = array[j];
307 |         array[j] = temp;
308 |     }
309 |     return array;
310 | }
311 | ```
312 | 
313 | KNN
314 | ----
315 | ```javascript
316 | const KNN = require('ml-knn');
317 | const csv = require('csvtojson');
318 | 
319 | let knn;
320 | 
321 | const csvFilePath = 'leaf.csv'; // Data
322 | const names = ['type', 'specimenNumber', 'eccentricity', 'aspectRatio', 'elongation', 'solidity', 'stochasticConvexity', 'isoperimetricFactor', 'maxIndetationDepth', 'lobedness', 'intensity', 'contrast', 'smoothness', 'thirdMoment', 'uniformity', 'entropy']; // For header
323 | 
324 | let seperationSize;
325 | let data = [], X = [], y = [];
326 | let trainingSetX = [], trainingSetY = [], testSetX = [], testSetY = [];
327 | 
328 | csv({noheader: true, headers: names})
329 |     .fromFile(csvFilePath)
330 |     .on('json', (jsonObj) => {
331 |         data.push(jsonObj); // Push each object to data Array
332 |     })
333 |     .on('done', (error) => {
334 |         seperationSize = 0.9 * data.length;
335 |         data = shuffleArray(data);
336 |         dressData();
337 |     });
338 | 
339 | function dressData() {
340 |     let types = new Set(); // To gather UNIQUE classes
341 |     data.forEach((row) => {
342 |         types.add(row.type);
343 |     });
344 |     typesArray = [...types]; // To save the different types of classes.
345 | 
346 |     data.forEach((row) => {
347 |         let rowArray, typeNumber;
348 |         rowArray = Object.keys(row).map(key => parseFloat(row[key])).slice(2, 16); // We don't use the 2 first elements, which are the type (i.e class) and the specimen number (i.e ID)
349 |         typeNumber = typesArray.indexOf(row.type); // Convert type(String) to type(Number)
350 | 
351 |         X.push(rowArray);
352 |         y.push(typeNumber);
353 |     });
354 | 
355 |     trainingSetX = X.slice(0, seperationSize);
356 |     trainingSetY = y.slice(0, seperationSize);
357 |     testSetX = X.slice(seperationSize);
358 |     testSetY = y.slice(seperationSize);
359 | 
360 |     train();
361 | }
362 | 
363 | function train() {
364 |     knn = new KNN(trainingSetX, trainingSetY, {k:5});
365 |     test();
366 | }
367 | 
368 | function test() {
369 |     const result = knn.predict(testSetX);
370 |     const testSetLength = testSetX.length
371 |     const predictionError = error(result, testSetY);
372 |     console.log(`Test Set Size = ${testSetLength} and number of Misclassifications = ${predictionError}`);
373 | }
374 | 
375 | function error(predicted, expected) {
376 |     let misclassifications = 0;
377 |     for (var index = 0; index < predicted.length; index++) {
378 |         console.log(`truth : ${expected[index]} and prediction : ${predicted[index]}`);
379 |         if (predicted[index] !== expected[index]) {
380 |             misclassifications++;
381 |         }
382 |     }
383 |     return misclassifications;
384 | }
385 | 
386 | function shuffleArray(array) {
387 |     for (var i = array.length - 1; i > 0; i--) {
388 |         var j = Math.floor(Math.random() * (i + 1));
389 |         var temp = array[i];
390 |         array[i] = array[j];
391 |         array[j] = temp;
392 |     }
393 |     return array;
394 | }
395 | ```
396 | 
397 | Naive bayes
398 | -----------
399 | ```javascript
400 | const Bayes = require('ml-naivebayes');
401 | const csv = require('csvtojson');
402 | 
403 | 
404 | bayes = new Bayes();
405 | 
406 | const csvFilePath = 'leaf.csv'; // Data
407 | const names = ['type', 'specimenNumber', 'eccentricity', 'aspectRatio', 'elongation', 'solidity', 'stochasticConvexity', 'isoperimetricFactor', 'maxIndetationDepth', 'lobedness', 'intensity', 'contrast', 'smoothness', 'thirdMoment', 'uniformity', 'entropy']; // For header
408 | 
409 | let seperationSize;
410 | let data = [], X = [], y = [];
411 | let trainingSetX = [], trainingSetY = [], testSetX = [], testSetY = [];
412 | 
413 | csv({noheader: true, headers: names})
414 |     .fromFile(csvFilePath)
415 |     .on('json', (jsonObj) => {
416 |         data.push(jsonObj); // Push each object to data Array
417 |     })
418 |     .on('done', (error) => {
419 |         seperationSize = 0.9 * data.length;
420 |         data = shuffleArray(data);
421 |         dressData();
422 |     });
423 | 
424 | function dressData() {
425 |     let types = new Set(); // To gather UNIQUE classes
426 |     data.forEach((row) => {
427 |         types.add(row.type);
428 |     });
429 |     typesArray = [...types]; // To save the different types of classes.
430 | 
431 |     data.forEach((row) => {
432 |         let rowArray, typeNumber;
433 |         rowArray = Object.keys(row).map(key => parseFloat(row[key])).slice(2, 16); // We don't use the 2 first elements, which are the type (i.e class) and the specimen number (i.e ID)
434 |         typeNumber = typesArray.indexOf(row.type); // Convert type(String) to type(Number)
435 | 
436 |         X.push(rowArray);
437 |         y.push(typeNumber);
438 |     });
439 | 
440 |     trainingSetX = X.slice(0, seperationSize);
441 |     trainingSetY = y.slice(0, seperationSize);
442 |     testSetX = X.slice(seperationSize);
443 |     testSetY = y.slice(seperationSize);
444 | 
445 |     train();
446 | }
447 | 
448 | function train() {
449 |     bayes.train(trainingSetX, trainingSetY);
450 |     test();
451 | }
452 | 
453 | function test() {
454 |     const result = bayes.predict(testSetX);
455 |     const testSetLength = testSetX.length
456 |     const predictionError = error(result, testSetY);
457 |     console.log(`Test Set Size = ${testSetLength} and number of Misclassifications = ${predictionError}`);
458 | }
459 | 
460 | function error(predicted, expected) {
461 |     let misclassifications = 0;
462 |     for (var index = 0; index < predicted.length; index++) {
463 |         console.log(`truth : ${expected[index]} and prediction : ${predicted[index]}`);
464 |         if (predicted[index] !== expected[index]) {
465 |             misclassifications++;
466 |         }
467 |     }
468 |     return misclassifications;
469 | }
470 | 
471 | function shuffleArray(array) {
472 |     for (var i = array.length - 1; i > 0; i--) {
473 |         var j = Math.floor(Math.random() * (i + 1));
474 |         var temp = array[i];
475 |         array[i] = array[j];
476 |         array[j] = temp;
477 |     }
478 |     return array;
479 | }
480 | ```
481 | 
482 | Logistic regression
483 | --------------------
484 | ```javascript
485 | const LogisticRegression = require('ml-logistic-regression');
486 | const csv = require('csvtojson');
487 | const {Matrix} = require('ml-matrix');
488 | 
489 | 
490 | let logreg = new LogisticRegression({numSteps: 5000, learningRate: 5e-3});
491 | 
492 | const csvFilePath = 'leaf.csv'; // Data
493 | const names = ['type', 'specimenNumber', 'eccentricity', 'aspectRatio', 'elongation', 'solidity', 'stochasticConvexity', 'isoperimetricFactor', 'maxIndetationDepth', 'lobedness', 'intensity', 'contrast', 'smoothness', 'thirdMoment', 'uniformity', 'entropy']; // For header
494 | 
495 | let seperationSize;
496 | let data = [], X = [], y = [];
497 | let trainingSetX = [], trainingSetY = [], testSetX = [], testSetY = [];
498 | 
499 | csv({noheader: true, headers: names})
500 |     .fromFile(csvFilePath)
501 |     .on('json', (jsonObj) => {
502 |         data.push(jsonObj); // Push each object to data Array
503 |     })
504 |     .on('done', (error) => {
505 |         seperationSize = 0.9 * data.length;
506 |         data = shuffleArray(data);
507 |         dressData();
508 |     });
509 | 
510 | function dressData() {
511 |     let types = new Set(); // To gather UNIQUE classes
512 |     data.forEach((row) => {
513 |         types.add(row.type);
514 |     });
515 |     typesArray = [...types]; // To save the different types of classes.
516 | 
517 |     data.forEach((row) => {
518 |         let rowArray, typeNumber;
519 |         rowArray = Object.keys(row).map(key => parseFloat(row[key])).slice(2, 16); // We don't use the 2 first elements, which are the type (i.e class) and the specimen number (i.e ID)
520 |         typeNumber = typesArray.indexOf(row.type); // Convert type(String) to type(Number)
521 | 
522 |         X.push(rowArray);
523 |         y.push(typeNumber);
524 |     });
525 | 
526 |     trainingSetX = new Matrix(X.slice(0, seperationSize));
527 |     trainingSetY = Matrix.columnVector(y.slice(0, seperationSize));
528 |     testSetX = new Matrix(X.slice(seperationSize));
529 |     testSetY = y.slice(seperationSize);
530 |     train();
531 | }
532 | 
533 | function train() {
534 |     logreg.train(trainingSetX, trainingSetY);
535 |     test();
536 | }
537 | 
538 | function test() {
539 |     const result = logreg.predict(new Matrix(testSetX));
540 |     const testSetLength = testSetX.length
541 |     const predictionError = error(result, testSetY);
542 |     console.log(`Test Set Size = ${testSetLength} and number of Misclassifications = ${predictionError}`);
543 | }
544 | 
545 | function error(predicted, expected) {
546 |     let misclassifications = 0;
547 |     for (var index = 0; index < predicted.length; index++) {
548 |         console.log(`truth : ${expected[index]} and prediction : ${predicted[index]}`);
549 |         if (predicted[index] !== expected[index]) {
550 |             misclassifications++;
551 |         }
552 |     }
553 |     return misclassifications;
554 | }
555 | 
556 | function shuffleArray(array) {
557 |     for (var i = array.length - 1; i > 0; i--) {
558 |         var j = Math.floor(Math.random() * (i + 1));
559 |         var temp = array[i];
560 |         array[i] = array[j];
561 |         array[j] = temp;
562 |     }
563 |     return array;
564 | }
565 | ```
566 | 
567 | Bonus : dimensionality reduction
568 | --------------------------------
569 | 
570 | Here, each data has 30 features. We can reduce the number of features of the dataset by using different algorithms (PCA, Isomap, etc.). The goal ? The fewer dimensions you will have, the faster the computation will be and if you reduce the dimensionality to 2 or 3, you can plot the data (it can be very practical).
571 | I give you an example with the PCA. Below, the source code using a PCA :
572 | ```javascript
573 | const KNN = require('ml-knn');
574 | const csv = require('csvtojson');
575 | const Matrix = require('ml-matrix');
576 | const PCA = require('ml-pca');
577 | 
578 | let usePCA = true;
579 | 
580 | let knn;
581 | 
582 | const csvFilePath = 'leaf.csv'; // Data
583 | const names = ['type', 'specimenNumber', 'eccentricity', 'aspectRatio', 'elongation', 'solidity', 'stochasticConvexity', 'isoperimetricFactor', 'maxIndetationDepth', 'lobedness', 'intensity', 'contrast', 'smoothness', 'thirdMoment', 'uniformity', 'entropy']; // For header
584 | 
585 | let seperationSize;
586 | let data = [], X = [], y = [];
587 | let trainingSetX = [], trainingSetY = [], testSetX = [], testSetY = [];
588 | 
589 | csv({noheader: true, headers: names})
590 |     .fromFile(csvFilePath)
591 |     .on('json', (jsonObj) => {
592 |         data.push(jsonObj); // Push each object to data Array
593 |     })
594 |     .on('done', (error) => {
595 |         seperationSize = 0.9 * data.length;
596 |         data = shuffleArray(data);
597 |         dressData();
598 |     });
599 | 
600 | function dressData() {
601 |     let types = new Set(); // To gather UNIQUE classes
602 |     data.forEach((row) => {
603 |         types.add(row.type);
604 |     });
605 |     typesArray = [...types]; // To save the different types of classes.
606 | 
607 |     data.forEach((row) => {
608 |         let rowArray, typeNumber;
609 |         rowArray = Object.keys(row).map(key => parseFloat(row[key])).slice(2, 16); // We don't use the 2 first elements, which are the type (i.e class) and the specimen number (i.e ID)
610 |         typeNumber = typesArray.indexOf(row.type); // Convert type(String) to type(Number)
611 | 
612 |         X.push(rowArray);
613 |         y.push(typeNumber);
614 |     });
615 | 
616 |     trainingSetX = X.slice(0, seperationSize);
617 |     trainingSetY = y.slice(0, seperationSize);
618 |     testSetX = X.slice(seperationSize);
619 |     testSetY = y.slice(seperationSize);
620 | 
621 |     train();
622 | }
623 | 
624 | function train() {
625 |     if (usePCA === true) {
626 |         trainingSetX = new Matrix(trainingSetX);
627 |         testSetX = new Matrix(testSetX);
628 |         var pca = new PCA(trainingSetX);
629 |         trainingSetX = pca.predict(trainingSetX, 2);
630 |         testSetX = pca.predict(testSetX, 2);
631 |     }
632 |     knn = new KNN(trainingSetX, trainingSetY, {k:5});
633 |     test();
634 | }
635 | 
636 | function test() {
637 |     const result = knn.predict(testSetX);
638 |     const testSetLength = testSetX.length
639 |     const predictionError = error(result, testSetY);
640 |     console.log(`Test Set Size = ${testSetLength} and number of Misclassifications = ${predictionError}`);
641 | }
642 | 
643 | function error(predicted, expected) {
644 |     let misclassifications = 0;
645 |     for (var index = 0; index < predicted.length; index++) {
646 |         console.log(`truth : ${expected[index]} and prediction : ${predicted[index]}`);
647 |         if (predicted[index] !== expected[index]) {
648 |             misclassifications++;
649 |         }
650 |     }
651 |     return misclassifications;
652 | }
653 | 
654 | function shuffleArray(array) {
655 |     for (var i = array.length - 1; i > 0; i--) {
656 |         var j = Math.floor(Math.random() * (i + 1));
657 |         var temp = array[i];
658 |         array[i] = array[j];
659 |         array[j] = temp;
660 |     }
661 |     return array;
662 | }
663 | ```
664 | 
665 | 
666 | Note :
667 | ======
668 | 
669 | Thanks to Abhishek Soni for the two tutorials https://hackernoon.com/machine-learning-with-javascript-part-1-9b97f3ed4fe5 and https://hackernoon.com/machine-learning-with-javascript-part-2-da994c17d483. My code reuse parts of the second tutorial (which is the solving of a classification problem with the method of KNN).
670 | 


--------------------------------------------------------------------------------
/examples/leafDataset/knn.js:
--------------------------------------------------------------------------------
 1 | const KNN = require('ml-knn');
 2 | const csv = require('csvtojson');
 3 | const Matrix = require('ml-matrix');
 4 | const PCA = require('ml-pca');
 5 | 
 6 | let usePCA = false;
 7 | 
 8 | let knn;
 9 | 
10 | const csvFilePath = 'leaf.csv'; // Data
11 | const names = ['type', 'specimenNumber', 'eccentricity', 'aspectRatio', 'elongation', 'solidity', 'stochasticConvexity', 'isoperimetricFactor', 'maxIndetationDepth', 'lobedness', 'intensity', 'contrast', 'smoothness', 'thirdMoment', 'uniformity', 'entropy']; // For header
12 | 
13 | let seperationSize;
14 | let data = [], X = [], y = [];
15 | let trainingSetX = [], trainingSetY = [], testSetX = [], testSetY = [];
16 | 
17 | csv({noheader: true, headers: names})
18 |     .fromFile(csvFilePath)
19 |     .on('json', (jsonObj) => {
20 |         data.push(jsonObj); // Push each object to data Array
21 |     })
22 |     .on('done', (error) => {
23 |         seperationSize = 0.9 * data.length;
24 |         data = shuffleArray(data);
25 |         dressData();
26 |     });
27 | 
28 | function dressData() {
29 |     let types = new Set(); // To gather UNIQUE classes
30 |     data.forEach((row) => {
31 |         types.add(row.type);
32 |     });
33 |     typesArray = [...types]; // To save the different types of classes.
34 | 
35 |     data.forEach((row) => {
36 |         let rowArray, typeNumber;
37 |         rowArray = Object.keys(row).map(key => parseFloat(row[key])).slice(2, 16); // We don't use the 2 first elements, which are the type (i.e class) and the specimen number (i.e ID)
38 |         typeNumber = typesArray.indexOf(row.type); // Convert type(String) to type(Number)
39 | 
40 |         X.push(rowArray);
41 |         y.push(typeNumber);
42 |     });
43 | 
44 |     trainingSetX = X.slice(0, seperationSize);
45 |     trainingSetY = y.slice(0, seperationSize);
46 |     testSetX = X.slice(seperationSize);
47 |     testSetY = y.slice(seperationSize);
48 | 
49 |     train();
50 | }
51 | 
52 | function train() {
53 |     if (usePCA === true) {
54 |         trainingSetX = new Matrix(trainingSetX);
55 |         testSetX = new Matrix(testSetX);
56 |         var pca = new PCA(trainingSetX);
57 |         trainingSetX = pca.predict(trainingSetX, 2);
58 |         testSetX = pca.predict(testSetX, 2);
59 |     }
60 |     knn = new KNN(trainingSetX, trainingSetY, {k:1});
61 |     test();
62 | }
63 | 
64 | function test() {
65 |     const result = knn.predict(testSetX);
66 |     const testSetLength = testSetX.length
67 |     const predictionError = error(result, testSetY);
68 |     console.log(`Test Set Size = ${testSetLength} and number of Misclassifications = ${predictionError}`);
69 | }
70 | 
71 | function error(predicted, expected) {
72 |     let misclassifications = 0;
73 |     for (var index = 0; index < predicted.length; index++) {
74 |         console.log(`truth : ${expected[index]} and prediction : ${predicted[index]}`);
75 |         if (predicted[index] !== expected[index]) {
76 |             misclassifications++;
77 |         }
78 |     }
79 |     return misclassifications;
80 | }
81 | 
82 | function shuffleArray(array) {
83 |     for (var i = array.length - 1; i > 0; i--) {
84 |         var j = Math.floor(Math.random() * (i + 1));
85 |         var temp = array[i];
86 |         array[i] = array[j];
87 |         array[j] = temp;
88 |     }
89 |     return array;
90 | }
91 | 


--------------------------------------------------------------------------------
/examples/leafDataset/leaf.csv:
--------------------------------------------------------------------------------
  1 | 1,1,0.72694,1.4742,0.32396,0.98535,1,0.83592,0.0046566,0.0039465,0.04779,0.12795,0.016108,0.0052323,0.00027477,1.1756
  2 | 1,2,0.74173,1.5257,0.36116,0.98152,0.99825,0.79867,0.0052423,0.0050016,0.02416,0.090476,0.0081195,0.002708,7.4846e-05,0.69659
  3 | 1,3,0.76722,1.5725,0.38998,0.97755,1,0.80812,0.0074573,0.010121,0.011897,0.057445,0.0032891,0.00092068,3.7886e-05,0.44348
  4 | 1,4,0.73797,1.4597,0.35376,0.97566,1,0.81697,0.0068768,0.0086068,0.01595,0.065491,0.0042707,0.0011544,6.6272e-05,0.58785
  5 | 1,5,0.82301,1.7707,0.44462,0.97698,1,0.75493,0.007428,0.010042,0.0079379,0.045339,0.0020514,0.00055986,2.3504e-05,0.34214
  6 | 1,6,0.72997,1.4892,0.34284,0.98755,1,0.84482,0.0049451,0.0044506,0.010487,0.058528,0.0034138,0.0011248,2.4798e-05,0.34068
  7 | 1,7,0.82063,1.7529,0.44458,0.97964,0.99649,0.7677,0.0059279,0.0063954,0.018375,0.080587,0.0064523,0.0022713,4.1495e-05,0.53904
  8 | 1,8,0.77982,1.6215,0.39222,0.98512,0.99825,0.80816,0.0050987,0.0047314,0.024875,0.089686,0.0079794,0.0024664,0.00014676,0.66975
  9 | 1,9,0.83089,1.8199,0.45693,0.9824,1,0.77106,0.0060055,0.006564,0.0072447,0.040616,0.0016469,0.00038812,3.2863e-05,0.33696
 10 | 1,10,0.90631,2.3906,0.58336,0.97683,0.99825,0.66419,0.0084019,0.012848,0.0070096,0.042347,0.0017901,0.00045889,2.8251e-05,0.28082
 11 | 1,11,0.7459,1.4927,0.34116,0.98296,1,0.83088,0.0055665,0.0056395,0.0057679,0.036511,0.0013313,0.00030872,3.1839e-05,0.25026
 12 | 1,12,0.79606,1.6934,0.43387,0.98181,1,0.76985,0.0077992,0.011071,0.013677,0.057832,0.0033334,0.00081648,0.00013855,0.49751
 13 | 2,1,0.93361,2.7582,0.64257,0.98346,1,0.59851,0.0055336,0.0055731,0.029712,0.089889,0.0080153,0.0020648,0.00023883,0.91499
 14 | 2,2,0.91186,2.4994,0.60323,0.983,1,0.64916,0.0061494,0.0068823,0.018887,0.072486,0.0052267,0.0014887,8.3271e-05,0.67811
 15 | 2,3,0.89063,2.2927,0.56667,0.98732,1,0.66427,0.0028365,0.0014643,0.029272,0.091328,0.0082717,0.0022383,0.00020166,0.87177
 16 | 2,4,0.86755,2.009,0.51464,0.98691,1,0.70277,0.0054439,0.0053937,0.030348,0.092063,0.0084044,0.0022541,0.00019854,0.94545
 17 | 2,5,0.91852,2.5247,0.61648,0.9787,1,0.63037,0.0050494,0.0046404,0.02309,0.082029,0.0066839,0.0018929,0.00012452,0.71713
 18 | 2,6,0.88795,2.2038,0.56218,0.97835,0.99825,0.64158,0.0059242,0.0063874,0.032722,0.092969,0.0085691,0.0021199,0.00027729,1.008
 19 | 2,7,0.85121,1.9548,0.4892,0.98622,1,0.70267,0.0039733,0.0028733,0.020258,0.070841,0.0049933,0.0012274,0.00014929,0.74174
 20 | 2,8,0.89084,2.2979,0.57815,0.97389,1,0.64598,0.015271,0.042443,0.028461,0.086477,0.0074228,0.0018832,0.00024345,0.91307
 21 | 2,9,0.93062,2.8973,0.65828,0.98182,1,0.5795,0.0064894,0.0076645,0.023606,0.072237,0.005191,0.0011217,0.00025558,0.90513
 22 | 2,10,0.84113,1.86,0.46549,0.99039,1,0.75976,0.0046759,0.0039793,0.062798,0.13234,0.017213,0.0044528,0.00065523,1.653
 23 | 3,1,0.70273,1.2099,0.36317,0.9211,0.98772,0.60555,0.023597,0.10134,0.089301,0.20088,0.038786,0.015895,0.0004049,1.5371
 24 | 3,2,0.66307,1.2065,0.32559,0.94952,0.99649,0.75954,0.013388,0.032621,0.021815,0.097143,0.0093485,0.0040284,3.6303e-05,0.5341
 25 | 3,3,0.61289,1.0991,0.33117,0.92405,0.98421,0.61661,0.025545,0.11877,0.054687,0.1606,0.025145,0.011672,0.00012121,1.1076
 26 | 3,4,0.70668,1.251,0.38111,0.94226,0.99825,0.6925,0.019432,0.068724,0.031587,0.11502,0.013056,0.0053112,8.6352e-05,0.72247
 27 | 3,5,0.66889,1.1435,0.3846,0.90355,0.99649,0.60571,0.028329,0.14606,0.057506,0.15931,0.024752,0.010304,0.00018533,1.1365
 28 | 3,6,0.50139,1.0066,0.29593,0.91585,0.99825,0.64029,0.021782,0.086347,0.054635,0.1598,0.024899,0.011106,0.00016162,1.0511
 29 | 3,7,0.60803,1.0646,0.3446,0.90487,0.99649,0.67517,0.031915,0.18538,0.06245,0.16411,0.026225,0.010602,0.00022964,1.2307
 30 | 3,8,0.56599,1.0427,0.35318,0.89086,0.99825,0.62068,0.032971,0.19785,0.026348,0.10589,0.011088,0.0046505,5.2046e-05,0.62671
 31 | 3,9,0.68605,1.1632,0.37528,0.91639,0.99825,0.61136,0.030339,0.16753,0.058401,0.16858,0.027633,0.012886,0.00015525,1.0815
 32 | 3,10,0.59571,1.084,0.32592,0.92056,0.99825,0.65698,0.024753,0.11151,0.042051,0.14722,0.021213,0.010881,6.468e-05,0.83221
 33 | 4,1,0.39432,1.1191,0.17272,0.96806,0.99298,0.73417,0.012499,0.028432,0.073063,0.13718,0.01847,0.0040597,0.00090977,1.8893
 34 | 4,2,0.62011,1.3298,0.2836,0.96952,0.99825,0.77871,0.0086562,0.013637,0.04722,0.1186,0.013872,0.0037549,0.00044188,1.2024
 35 | 4,3,0.73429,1.5069,0.40827,0.94518,1,0.62558,0.014987,0.040879,0.038113,0.092188,0.0084269,0.0017773,0.00047378,1.335
 36 | 4,4,0.64161,1.3004,0.31556,0.93324,0.97895,0.65487,0.017963,0.058729,0.030897,0.08221,0.0067132,0.0014678,0.00029878,1.2359
 37 | 4,5,0.64056,1.3166,0.28616,0.97053,0.99649,0.77183,0.0091632,0.015281,0.020413,0.065575,0.0042817,0.00089749,0.00020433,0.86218
 38 | 4,6,0.74242,1.5529,0.38804,0.95593,0.99474,0.68188,0.012904,0.030306,0.020679,0.064242,0.0041101,0.00079505,0.00026812,0.87687
 39 | 4,7,0.39289,1.1286,0.17039,0.96405,1,0.79407,0.011761,0.025176,0.025409,0.078211,0.0060798,0.0015299,0.00015762,1.0326
 40 | 4,8,0.53017,1.2503,0.23254,0.97337,1,0.82703,0.0066263,0.0079913,0.048541,0.10941,0.011828,0.0027809,0.00048839,1.5452
 41 | 5,1,0.87844,1.8096,0.63151,0.83923,0.83684,0.37688,0.043563,0.34539,0.045468,0.12554,0.015515,0.0051589,0.00022842,1.1374
 42 | 5,2,0.88075,1.736,0.58345,0.83383,0.91754,0.41551,0.040582,0.29973,0.035786,0.10047,0.0099932,0.002759,0.00025802,1.0952
 43 | 5,3,0.86545,1.8803,0.62039,0.82443,0.85439,0.33077,0.047,0.40204,0.039518,0.1157,0.01321,0.0042406,0.00020084,1.0136
 44 | 5,4,0.93671,2.4151,0.7298,0.81793,0.86491,0.33439,0.080539,1.1805,0.048722,0.12051,0.014314,0.0039983,0.00037216,1.3083
 45 | 5,5,0.92676,2.222,0.6558,0.82432,0.89474,0.35618,0.038012,0.26298,0.059981,0.14901,0.021723,0.0075251,0.00028398,1.353
 46 | 5,6,0.90944,2.1139,0.69555,0.80873,0.85263,0.31039,0.047267,0.40663,0.039446,0.10517,0.010939,0.0028913,0.00029366,1.1602
 47 | 5,7,0.89493,2.0494,0.64925,0.81682,0.83509,0.35318,0.049955,0.45418,0.041808,0.12036,0.01428,0.0048007,0.00017096,1.0962
 48 | 5,8,0.93613,2.3898,0.67175,0.84434,0.90526,0.37007,0.039277,0.28077,0.048447,0.11955,0.014091,0.0037887,0.00042459,1.2593
 49 | 5,9,0.88172,1.774,0.63974,0.8499,0.87368,0.34354,0.051776,0.4879,0.06103,0.1519,0.022554,0.0080813,0.00025039,1.3805
 50 | 5,10,0.86224,1.7434,0.62464,0.85754,0.89298,0.32238,0.050792,0.46952,0.064428,0.16335,0.025989,0.010189,0.00023873,1.3343
 51 | 5,11,0.82268,1.6657,0.59909,0.84493,0.85614,0.32056,0.048803,0.43348,0.079282,0.18571,0.033338,0.013523,0.00027502,1.5377
 52 | 5,12,0.82135,1.6035,0.6025,0.84981,0.87544,0.31304,0.056079,0.57236,0.069239,0.16319,0.02594,0.0092918,0.00032134,1.4686
 53 | 6,1,0.46348,1.0544,0.72231,0.63754,0.47544,0.20039,0.14864,4.0212,0.010917,0.053494,0.0028534,0.00087543,2.9594e-05,0.53199
 54 | 6,2,0.50784,1.0544,0.623,0.73625,0.59649,0.27687,0.13296,3.2176,0.014816,0.067044,0.0044748,0.0013737,3.792e-05,0.52114
 55 | 6,3,0.47728,1.2115,0.68343,0.76364,0.56667,0.23896,0.10747,2.1022,0.024036,0.080623,0.0064581,0.0018312,0.00010648,0.90702
 56 | 6,4,0.52744,1.0419,0.51662,0.77201,0.84211,0.33784,0.11438,2.3811,0.0254,0.082047,0.0066868,0.0017982,0.00012196,0.944
 57 | 6,5,0.50964,1.2067,0.70618,0.72175,0.56491,0.21009,0.11264,2.3093,0.021181,0.080306,0.0064077,0.0020323,6.5828e-05,0.73274
 58 | 6,6,0.63965,1.2323,0.60663,0.77037,0.62105,0.24135,0.12438,2.8155,0.025438,0.096215,0.0091723,0.0034208,5.1566e-05,0.75194
 59 | 6,7,0.71847,1.2291,0.693,0.75574,0.66667,0.26004,0.098823,1.7774,0.021217,0.073201,0.0053299,0.0013612,0.00012452,0.79862
 60 | 6,8,0.47815,1.1032,0.52998,0.81654,0.8193,0.33753,0.11547,2.4268,0.006859,0.042071,0.0017668,0.00047564,1.4404e-05,0.30832
 61 | 7,1,0.84886,1.7011,0.5348,0.81465,0.97018,0.37276,0.035424,0.22838,0.038846,0.1192,0.014009,0.0057424,0.00010856,1.2269
 62 | 7,2,0.83755,1.5634,0.53329,0.76697,0.95088,0.27825,0.039592,0.28528,0.023152,0.087838,0.0076564,0.0028187,5.6309e-05,0.8079
 63 | 7,3,0.84315,1.8134,0.54611,0.84406,0.96667,0.49437,0.038509,0.2699,0.054653,0.14411,0.020345,0.0082654,0.00016275,1.495
 64 | 7,4,0.82602,1.6637,0.49244,0.8862,0.99298,0.57136,0.022152,0.089306,0.034444,0.1012,0.010138,0.0030708,0.00016513,1.1047
 65 | 7,5,0.86486,1.7986,0.55606,0.87745,0.97018,0.52669,0.029555,0.15898,0.027883,0.092866,0.0085504,0.0031577,0.00012091,0.9961
 66 | 7,6,0.84563,1.7418,0.49818,0.92748,1,0.62301,0.014708,0.039374,0.025028,0.077667,0.0059959,0.002105,0.00029619,1.1899
 67 | 7,7,0.87163,2.0659,0.55339,0.94868,1,0.60231,0.031671,0.18255,0.049706,0.1344,0.017742,0.0066921,0.00016133,1.4016
 68 | 7,8,0.86587,1.9999,0.52725,0.97215,0.99474,0.67283,0.0092956,0.015726,0.021192,0.073946,0.0054383,0.0014673,0.00015488,0.78201
 69 | 7,9,0.82373,1.6421,0.47458,0.85777,0.98421,0.50374,0.020049,0.073157,0.018698,0.075567,0.005678,0.0023517,8.791e-05,0.7658
 70 | 7,10,0.84285,1.8572,0.51931,0.91339,0.97544,0.60526,0.028734,0.15027,0.038283,0.1156,0.013186,0.0048867,0.00012558,1.1185
 71 | 8,1,0.98502,6.0596,0.83612,0.97628,0.98772,0.31546,0.02322,0.098133,0.03683,0.10661,0.011238,0.0032908,0.00022206,1.0185
 72 | 8,2,0.98487,6.0229,0.83628,0.96522,1,0.29509,0.024667,0.11074,0.0303,0.097095,0.0093394,0.0027833,0.00018341,0.85025
 73 | 8,3,0.98853,6.628,0.84928,0.96861,0.97719,0.27717,0.029032,0.1534,0.032722,0.10283,0.010462,0.0033387,0.00013116,0.94985
 74 | 8,4,0.98326,5.8308,0.82878,0.96489,1,0.30777,0.014672,0.039178,0.019278,0.073118,0.0053178,0.0014657,9.7882e-05,0.66403
 75 | 8,5,0.98592,6.0269,0.83884,0.96027,0.96842,0.28345,0.025004,0.11379,0.021226,0.080869,0.0064972,0.001957,9.1646e-05,0.64839
 76 | 8,6,0.98695,6.2904,0.84124,0.97085,1,0.29288,0.023221,0.09814,0.025859,0.089821,0.0080032,0.0024919,9.7883e-05,0.8112
 77 | 8,7,0.98717,6.5173,0.84726,0.96846,1,0.28899,0.022056,0.08854,0.042124,0.11842,0.013829,0.0043818,0.00020215,1.098
 78 | 8,8,0.98361,5.8719,0.82847,0.95488,0.98421,0.32134,0.040164,0.29359,0.039243,0.11622,0.013327,0.0043872,0.00019465,1.0006
 79 | 8,9,0.98643,6.1273,0.84737,0.95274,0.89649,0.29982,0.019421,0.068643,0.028986,0.095642,0.0090645,0.0027526,0.00014425,0.82529
 80 | 8,10,0.986,6.1763,0.83942,0.97837,0.99298,0.31607,0.011769,0.025208,0.051048,0.13524,0.017961,0.0060176,0.0003128,1.166
 81 | 8,11,0.98824,7.142,0.86069,0.96683,0.99474,0.26951,0.021697,0.08568,0.024361,0.087108,0.0075307,0.0023055,9.7934e-05,0.7396
 82 | 9,1,0.47821,1.2059,0.34976,0.90378,0.99649,0.51298,0.030166,0.16561,0.082509,0.17347,0.029213,0.0093049,0.00067871,1.5956
 83 | 9,2,0.53941,1.278,0.32286,0.91793,0.97368,0.63545,0.034884,0.22147,0.060967,0.15124,0.022362,0.0076389,0.00043953,1.2541
 84 | 9,3,0.54833,1.2733,0.31093,0.93409,0.99825,0.59707,0.017966,0.058744,0.043685,0.12505,0.015397,0.0051037,0.00042441,0.94617
 85 | 9,4,0.58938,1.3232,0.40145,0.89009,0.98421,0.43977,0.047907,0.4177,0.11488,0.19211,0.035592,0.0091838,0.0015424,2.1004
 86 | 9,5,0.64803,1.392,0.41362,0.89752,0.96316,0.50892,0.026737,0.1301,0.058284,0.15033,0.022099,0.0078269,0.00036426,1.184
 87 | 9,6,0.54029,1.2759,0.36992,0.88831,0.97719,0.54283,0.028138,0.1441,0.051554,0.14256,0.019917,0.0074501,0.00020354,1.1427
 88 | 9,7,0.4309,1.2825,0.38769,0.85341,0.96667,0.46204,0.052403,0.49979,0.067688,0.16198,0.025566,0.0091042,0.00043085,1.3596
 89 | 9,8,0.4659,1.2195,0.27095,0.94433,1,0.66599,0.013322,0.032299,0.024332,0.10405,0.01071,0.0047135,4.6307e-05,0.56447
 90 | 9,9,0.4867,1.2165,0.37691,0.90161,0.99474,0.54006,0.032181,0.18848,0.098716,0.18377,0.03267,0.009529,0.00095661,1.9018
 91 | 9,10,0.46098,1.2199,0.32967,0.90459,1,0.56329,0.027061,0.13328,0.092156,0.18249,0.03223,0.010076,0.00090962,1.7156
 92 | 9,11,0.73086,1.5285,0.46455,0.87291,0.97193,0.4865,0.034289,0.21399,0.02898,0.10191,0.01028,0.0035859,0.00010051,0.76676
 93 | 9,12,0.51577,1.2902,0.32454,0.9208,0.99474,0.62017,0.023879,0.10378,0.047501,0.12228,0.014732,0.0042129,0.00056945,1.1232
 94 | 9,13,0.59837,1.3539,0.35704,0.93647,0.99825,0.64448,0.019818,0.071484,0.045326,0.13137,0.016966,0.0059817,0.0003532,0.93682
 95 | 9,14,0.55112,1.2821,0.39683,0.88375,0.99649,0.49492,0.030576,0.17015,0.062015,0.14621,0.02093,0.0063685,0.00067747,1.2659
 96 | 10,1,0.44941,1.1374,0.38458,0.90736,0.97368,0.54595,0.047809,0.41601,0.11385,0.18608,0.033467,0.0082395,0.0012682,2.2422
 97 | 10,2,0.46727,1.1876,0.36748,0.91542,0.97719,0.57995,0.051335,0.47962,0.13764,0.18732,0.033899,0.0061636,0.0029358,2.5551
 98 | 10,3,0.55362,1.2438,0.39517,0.91092,0.97193,0.51113,0.046323,0.39055,0.083154,0.17465,0.0296,0.0094574,0.00074607,1.5807
 99 | 10,4,0.32175,1.1782,0.32151,0.91803,0.97544,0.58719,0.045933,0.384,0.076818,0.14622,0.020931,0.0048174,0.0014091,1.6749
100 | 10,5,0.64414,1.37,0.42697,0.91457,0.96491,0.44764,0.055411,0.55882,0.1203,0.19149,0.035373,0.0084315,0.0018137,2.2137
101 | 10,6,0.55977,1.3442,0.34301,0.9298,0.97544,0.57879,0.053564,0.52218,0.14905,0.25543,0.061249,0.02381,0.00059701,2.413
102 | 10,7,0.50114,1.2499,0.30101,0.94477,0.9807,0.6743,0.0494,0.44415,0.096252,0.23034,0.050382,0.025877,0.00026934,1.4463
103 | 10,8,0.23041,1.1294,0.25496,0.94201,0.97368,0.6084,0.040722,0.3018,0.10957,0.23571,0.052636,0.02418,0.00042353,1.6648
104 | 10,9,0.46094,1.1366,0.22905,0.95565,0.97544,0.71234,0.038286,0.26678,0.11753,0.22041,0.046329,0.016554,0.00064766,1.9866
105 | 10,10,0.54913,1.2769,0.31283,0.93889,0.99123,0.5933,0.041539,0.31404,0.15188,0.2605,0.06355,0.02379,0.000843,2.203
106 | 10,11,0.32168,1.1127,0.27055,0.95426,0.99298,0.70006,0.05426,0.53584,0.10967,0.21507,0.044212,0.016252,0.00061195,1.8452
107 | 10,12,0.38564,1.0471,0.23328,0.95176,0.97368,0.67107,0.050832,0.47028,0.13563,0.24744,0.057693,0.022,0.00086835,1.9757
108 | 10,13,0.36127,1.0573,0.26213,0.93923,0.96667,0.59302,0.04962,0.44811,0.19067,0.28081,0.073089,0.023825,0.0012501,2.6604
109 | 11,1,0.51247,1.1116,0.65626,0.57724,0.59298,0.16867,0.11187,2.2776,0.016001,0.061238,0.0037361,0.00087861,0.00010284,0.6508
110 | 11,2,0.54893,1.1111,0.63983,0.56623,0.6,0.15743,0.13081,3.1143,0.021231,0.079722,0.0063155,0.0019123,7.3919e-05,0.71949
111 | 11,3,0.43425,1.095,0.68828,0.52398,0.54211,0.15396,0.13761,3.4462,0.021147,0.073202,0.0053299,0.00134,0.00016376,0.75496
112 | 11,4,0.38501,1.0656,0.63042,0.51223,0.59123,0.13705,0.12292,2.7498,0.033373,0.098907,0.0096879,0.0027872,0.00021426,1.0015
113 | 11,5,0.26758,1.1316,0.60128,0.54301,0.77368,0.20311,0.15354,4.2904,0.017945,0.07145,0.0050792,0.0014652,7.2478e-05,0.63273
114 | 11,6,0.24465,1.047,0.60511,0.56524,0.79474,0.21788,0.12522,2.854,0.037595,0.127,0.015874,0.006587,0.00010798,0.8331
115 | 11,7,0.39092,1.087,0.68174,0.50961,0.6614,0.15361,0.14082,3.6093,0.028638,0.089135,0.0078824,0.0021177,0.00021044,0.90082
116 | 11,8,0.4042,1.0965,0.65899,0.52833,0.68421,0.1771,0.13017,3.0837,0.029057,0.096836,0.0092902,0.0028927,0.0001265,0.82383
117 | 11,9,0.50692,1.127,0.67203,0.53024,0.75263,0.16792,0.13006,3.0788,0.015279,0.057592,0.0033059,0.00072847,0.00010983,0.67289
118 | 11,10,0.47565,1.0656,0.69172,0.5233,0.49649,0.14133,0.12987,3.0697,0.023977,0.0842,0.0070397,0.0020845,0.00011284,0.77399
119 | 11,11,0.52382,1.1117,0.67175,0.54701,0.62982,0.15157,0.13674,3.4028,0.026434,0.085792,0.0073064,0.0021373,0.00016583,0.90513
120 | 11,12,0.36462,1.0811,0.67755,0.49042,0.68772,0.14118,0.1243,2.8118,0.037866,0.11692,0.013485,0.0046475,0.0001769,0.9229
121 | 11,13,0.52212,1.1191,0.70988,0.50678,0.64912,0.1412,0.13192,3.1674,0.025478,0.085964,0.0073356,0.0021793,0.0001485,0.82809
122 | 11,14,0.38203,1.0405,0.6901,0.48549,0.63684,0.13165,0.11852,2.5565,0.027997,0.093312,0.008632,0.0026591,0.00012469,0.84994
123 | 11,15,0.27123,1.096,0.68075,0.49446,0.53684,0.13088,0.12815,2.9887,0.021943,0.072882,0.0052838,0.0013391,0.00022058,0.80402
124 | 11,16,0.44882,1.0118,0.6301,0.57134,0.81053,0.16187,0.11115,2.2486,0.027309,0.088889,0.0078393,0.0022734,0.00017545,0.86
125 | 12,1,0.88374,2.4765,0.61389,0.8581,0.98246,0.40448,0.054511,0.54081,0.06489,0.14657,0.021032,0.0067378,0.00037204,1.6461
126 | 12,2,0.91579,2.5959,0.63023,0.94977,1,0.4976,0.024404,0.10839,0.067044,0.1528,0.022814,0.0075708,0.00031405,1.6221
127 | 12,3,0.82866,1.9848,0.50917,0.9418,0.99825,0.55942,0.025524,0.11857,0.080103,0.16692,0.027107,0.0086548,0.0004265,1.8038
128 | 12,4,0.87863,2.4734,0.60898,0.87869,0.99123,0.45758,0.056994,0.59119,0.076471,0.16369,0.026095,0.0087482,0.00034348,1.8363
129 | 12,5,0.92876,2.8218,0.66081,0.93335,0.97544,0.43699,0.046152,0.38767,0.050806,0.13408,0.017661,0.0064421,0.00017937,1.4027
130 | 12,6,0.91093,2.6184,0.63236,0.92868,0.99123,0.51584,0.021091,0.080956,0.05094,0.12676,0.015815,0.0047138,0.00031804,1.3323
131 | 12,7,0.92479,2.5638,0.62255,0.96226,0.97895,0.57504,0.0097927,0.017453,0.026156,0.10029,0.009958,0.0040505,4.4661e-05,0.77718
132 | 12,8,0.90037,2.3499,0.60125,0.93713,0.98421,0.46172,0.03166,0.18242,0.096235,0.18081,0.031658,0.0105,0.00050824,2.3052
133 | 12,9,0.8728,2.3151,0.58751,0.90716,0.98772,0.45257,0.058029,0.61287,0.07597,0.17378,0.029313,0.011448,0.00025642,1.6895
134 | 12,10,0.93377,2.8883,0.66665,0.93026,0.96316,0.43032,0.03571,0.23209,0.047675,0.12917,0.016412,0.0057866,0.00016113,1.321
135 | 12,11,0.88185,2.1732,0.57028,0.93013,0.99825,0.45014,0.049062,0.43809,0.072246,0.16601,0.02682,0.010288,0.0002503,1.7405
136 | 12,12,0.87259,2.1845,0.56922,0.93969,0.99474,0.52611,0.044598,0.36199,0.060388,0.14227,0.019839,0.0062705,0.00031494,1.4969
137 | 13,1,0.71763,1.5041,0.33723,0.98803,1,0.82342,0.0040421,0.0029736,0.050064,0.12542,0.015486,0.0046629,0.00027091,1.3598
138 | 13,2,0.73281,1.5437,0.36231,0.98616,1,0.79359,0.0052593,0.0050342,0.046731,0.1203,0.014265,0.0041563,0.00026442,1.2681
139 | 13,3,0.75285,1.5603,0.38474,0.98701,0.99649,0.78037,0.004611,0.0038695,0.046705,0.11723,0.013556,0.0038164,0.00028371,1.3539
140 | 13,4,0.73935,1.5319,0.34987,0.98479,1,0.81067,0.0078079,0.011095,0.027888,0.11472,0.01299,0.0060166,5.0098e-05,0.59895
141 | 13,5,0.63969,1.3612,0.28845,0.98738,1,0.82125,0.0042318,0.0032593,0.080446,0.15813,0.024394,0.0066506,0.00067629,1.8274
142 | 13,6,0.62033,1.3105,0.26312,0.98535,1,0.82478,0.0049896,0.0045311,0.070042,0.14655,0.021025,0.0058018,0.00053019,1.705
143 | 13,7,0.64536,1.3743,0.28162,0.98646,1,0.84494,0.005427,0.0053602,0.026853,0.085949,0.007333,0.0020334,0.00012813,0.94296
144 | 13,8,0.70115,1.4925,0.34048,0.98262,1,0.78473,0.0061511,0.0068862,0.08405,0.17102,0.028418,0.0087725,0.00052855,1.7782
145 | 13,9,0.70275,1.5016,0.34772,0.98776,0.99825,0.80848,0.0037534,0.002564,0.055512,0.12613,0.015659,0.004022,0.00050106,1.4395
146 | 13,10,0.69686,1.4507,0.33575,0.9851,1,0.80461,0.0036582,0.0024356,0.038211,0.10889,0.011717,0.003581,0.00016657,1.1363
147 | 13,11,0.66351,1.4024,0.31179,0.98529,1,0.81091,0.0068704,0.0085907,0.068206,0.13895,0.01894,0.0049098,0.00062422,1.7727
148 | 13,12,0.67362,1.3988,0.30608,0.98301,1,0.79998,0.0087621,0.013973,0.1196,0.17804,0.030726,0.0063197,0.0016787,2.5094
149 | 13,13,0.64314,1.3574,0.28063,0.98631,0.99825,0.81381,0.0052732,0.0050608,0.058531,0.13108,0.016892,0.0044631,0.00048306,1.5136
150 | 14,1,0.91298,2.455,0.61097,0.96979,0.99825,0.5621,0.0095958,0.016758,0.090447,0.1871,0.033823,0.0115,0.000564,1.6879
151 | 14,2,0.9136,2.5213,0.61341,0.97227,1,0.56248,0.0078474,0.011208,0.068427,0.14325,0.020108,0.0055763,0.00057194,1.7187
152 | 14,3,0.91355,2.457,0.61108,0.97034,0.97544,0.5645,0.0077029,0.010799,0.092547,0.18786,0.034088,0.011708,0.00046097,1.8501
153 | 14,4,0.91378,2.4132,0.60362,0.96618,0.99649,0.56153,0.0091377,0.015197,0.10604,0.20316,0.039637,0.013227,0.00072984,1.8526
154 | 14,5,0.85897,1.9723,0.51434,0.97407,0.99825,0.6792,0.0068392,0.0085129,0.046518,0.13848,0.018816,0.0073898,0.00015454,1.014
155 | 14,6,0.91296,2.4862,0.62315,0.96188,1,0.51041,0.010684,0.020775,0.064539,0.13678,0.018365,0.0048636,0.00052757,1.6875
156 | 14,7,0.93802,2.8087,0.65774,0.97091,1,0.58194,0.0067321,0.0082485,0.052218,0.13491,0.017874,0.0057374,0.00030246,1.2161
157 | 14,8,0.94058,3.0366,0.69248,0.93952,0.99474,0.42979,0.031219,0.17738,0.038691,0.1076,0.011445,0.0032156,0.00024816,1.0843
158 | 14,9,0.89672,2.2534,0.57689,0.96488,1,0.57906,0.0081422,0.012066,0.062778,0.1512,0.022351,0.0078553,0.00031192,1.4575
159 | 14,10,0.89507,2.2045,0.55605,0.97701,0.99649,0.64214,0.0053433,0.0051962,0.102,0.19598,0.036988,0.012562,0.00048437,2.0484
160 | 14,11,0.88246,2.1,0.5486,0.96929,0.99825,0.61897,0.0073021,0.0097043,0.10152,0.19717,0.03742,0.012995,0.00049125,2.0094
161 | 14,12,0.90831,2.4038,0.59263,0.97688,0.99825,0.59259,0.0064679,0.0076137,0.063688,0.1472,0.021208,0.0067017,0.00032702,1.5332
162 | 15,1,0.4132,1.0384,0.48465,0.78118,0.87018,0.30478,0.080722,1.1859,0.047303,0.12619,0.015674,0.0058992,0.00019037,1.489
163 | 15,2,0.56324,1.1928,0.474,0.77564,0.93333,0.34791,0.069404,0.87668,0.10401,0.17923,0.031125,0.0094494,0.00074739,2.4883
164 | 15,3,0.48967,1.1232,0.46175,0.80214,0.9,0.31034,0.073462,0.9822,0.10125,0.17926,0.031133,0.010102,0.00068131,2.4808
165 | 15,4,0.60319,1.088,0.51739,0.7638,0.90175,0.23802,0.077022,1.0797,0.10994,0.17976,0.031303,0.0088846,0.00087888,2.6613
166 | 15,5,0.53622,1.2341,0.44979,0.7882,0.94386,0.38426,0.072253,0.95013,0.042992,0.12337,0.014993,0.0056501,0.00013103,1.2646
167 | 15,6,0.37902,1.0738,0.5242,0.71586,0.85439,0.26309,0.065312,0.77635,0.067161,0.14275,0.019969,0.0059582,0.00040634,1.8527
168 | 15,7,0.53693,1.0751,0.55667,0.74395,0.77544,0.27241,0.080278,1.1729,0.091322,0.16265,0.025773,0.007318,0.00080697,2.2704
169 | 15,8,0.51153,1.0767,0.47053,0.77952,0.90702,0.28838,0.076992,1.0789,0.084711,0.17213,0.028776,0.0099656,0.00039644,2.0412
170 | 15,9,0.44271,1.0836,0.49548,0.74434,0.87018,0.26478,0.079481,1.1497,0.10151,0.1726,0.028929,0.0076406,0.00084487,2.389
171 | 15,10,0.30574,1.1176,0.58847,0.66502,0.84737,0.24074,0.10064,1.8435,0.074621,0.14273,0.019966,0.0054026,0.00062391,2.1403
172 | 22,1,0.92522,2.695,0.63571,0.92994,0.99298,0.505,0.040707,0.30158,0.055742,0.15592,0.023734,0.0096617,0.00030237,1.0351
173 | 22,2,0.90557,2.3423,0.58487,0.95943,0.97368,0.55537,0.023542,0.10087,0.045897,0.13433,0.017724,0.0066757,0.00024528,0.9943
174 | 22,3,0.91885,2.638,0.62859,0.96692,0.99123,0.52269,0.020081,0.07339,0.054145,0.1674,0.027257,0.013322,0.00016644,0.8989
175 | 22,4,0.88965,2.3863,0.59051,0.90564,0.98421,0.5136,0.048934,0.43581,0.062886,0.17454,0.029562,0.013364,0.00026357,1.0734
176 | 22,5,0.9456,3.1174,0.70912,0.79489,0.89298,0.33915,0.13457,3.2956,0.052685,0.14875,0.021648,0.008569,0.00025614,1.0503
177 | 22,6,0.91747,2.7054,0.64091,0.87687,0.98947,0.42531,0.089364,1.4534,0.042216,0.12098,0.014424,0.0047692,0.0003061,1.0097
178 | 22,7,0.86049,2.0997,0.532,0.9358,0.98772,0.5343,0.033994,0.21032,0.040254,0.13592,0.01814,0.0080226,0.00010143,0.82059
179 | 22,8,0.89808,2.4056,0.59137,0.93186,0.98596,0.51229,0.029398,0.15729,0.044185,0.15479,0.0234,0.012336,7.6339e-05,0.78887
180 | 22,9,0.86988,2.1277,0.53926,0.96257,1,0.64927,0.027899,0.14166,0.026849,0.096968,0.0093152,0.0032376,0.00010402,0.7271
181 | 22,10,0.86675,2.0417,0.52999,0.94477,0.98947,0.55141,0.028749,0.15043,0.031023,0.12007,0.014211,0.0064538,6.947e-05,0.64966
182 | 22,11,0.96652,3.8162,0.74273,0.82566,0.94737,0.32925,0.1446,3.8052,0.022715,0.10148,0.010193,0.0046292,4.4517e-05,0.52175
183 | 22,12,0.93396,2.9367,0.66689,0.92543,0.99474,0.45676,0.030052,0.16437,0.015159,0.071198,0.0050436,0.0017096,4.8527e-05,0.46762
184 | 23,1,0.61823,1.2435,0.33084,0.94147,0.9614,0.64089,0.024255,0.10708,0.017543,0.09745,0.0094072,0.0051286,1.2841e-05,0.37324
185 | 23,2,0.61717,1.2296,0.34897,0.93189,0.97368,0.60176,0.025375,0.11719,0.012906,0.070197,0.0049035,0.0019323,2.4448e-05,0.37728
186 | 23,3,0.62404,1.1599,0.47899,0.91562,0.85088,0.45154,0.097205,1.7197,0.013983,0.076408,0.0058043,0.0024109,2.5646e-05,0.36029
187 | 23,4,0.48229,1.1574,0.42617,0.91741,0.92456,0.46699,0.063396,0.73147,0.026522,0.105,0.010904,0.0044681,6.7833e-05,0.63468
188 | 23,5,0.5684,1.1876,0.48001,0.91247,0.79474,0.3885,0.089162,1.4469,0.027113,0.10182,0.010261,0.0037888,0.00010673,0.65567
189 | 23,6,0.62733,1.2441,0.50214,0.90049,0.86667,0.47758,0.092591,1.5603,0.02034,0.091645,0.0083289,0.0033696,4.9071e-05,0.49195
190 | 23,7,0.58637,1.1419,0.30339,0.93305,0.92105,0.57323,0.041282,0.31016,0.022886,0.093704,0.0087041,0.0032291,0.00010441,0.54481
191 | 23,8,0.50771,1.1084,0.25071,0.93529,0.98596,0.64927,0.02595,0.12256,0.013493,0.073509,0.0053745,0.0021246,3.1877e-05,0.3475
192 | 23,9,0.6018,1.2254,0.41835,0.94269,0.95263,0.54871,0.058835,0.62999,0.016202,0.074883,0.0055761,0.0018797,7.1264e-05,0.43769
193 | 23,10,0.63548,1.202,0.47365,0.89373,0.86842,0.44347,0.077616,1.0964,0.022403,0.10386,0.010672,0.005084,3.1292e-05,0.48911
194 | 23,11,0.65048,1.2354,0.44987,0.90929,0.89123,0.57594,0.070699,0.9097,0.0093349,0.0592,0.0034924,0.0013047,2.2232e-05,0.26169
195 | 24,1,0.64604,1.4683,0.35457,0.9588,0.99123,0.70641,0.02021,0.074334,0.059026,0.12721,0.015925,0.0037362,0.0012496,1.3596
196 | 24,2,0.55457,1.2955,0.25519,0.97441,0.99825,0.7824,0.012063,0.026483,0.020129,0.076722,0.0058518,0.0016389,0.00012751,0.60629
197 | 24,3,0.4443,1.2035,0.19432,0.98319,1,0.833,0.010317,0.019373,0.029059,0.10265,0.010427,0.003627,0.00011748,0.73255
198 | 24,4,0.6368,1.4266,0.29976,0.9747,0.99825,0.78457,0.013385,0.032608,0.028272,0.084495,0.0070888,0.0017155,0.00031351,0.913
199 | 24,5,0.64168,1.4853,0.3362,0.95562,0.99825,0.72952,0.017439,0.05535,0.02413,0.074641,0.0055403,0.001194,0.00031921,0.82279
200 | 24,6,0.58232,1.343,0.29271,0.95055,0.99825,0.71499,0.026876,0.13146,0.068134,0.13204,0.017136,0.0036798,0.0014697,1.6315
201 | 24,7,0.44567,1.2107,0.20257,0.97104,0.99123,0.80094,0.016177,0.047628,0.020632,0.074115,0.0054631,0.0013865,0.00016174,0.66423
202 | 24,8,0.67925,1.487,0.33465,0.97307,0.98246,0.7713,0.012399,0.02798,0.0075996,0.044132,0.0019439,0.00049059,3.6857e-05,0.28857
203 | 24,9,0.11708,1.0944,0.1745,0.97545,1,0.81957,0.016114,0.047259,0.033007,0.087092,0.0075278,0.001568,0.0006066,1.0292
204 | 24,10,0.53283,1.3078,0.25925,0.96727,1,0.77065,0.017745,0.057312,0.05341,0.12259,0.014806,0.0035923,0.0012358,1.1886
205 | 24,11,0.52828,1.2496,0.2238,0.98317,1,0.82495,0.010372,0.01958,0.047482,0.10274,0.010444,0.0019794,0.0011976,1.341
206 | 24,12,0.35236,1.146,0.20076,0.96876,0.99649,0.71333,0.017729,0.057204,0.033249,0.099068,0.0097191,0.002653,0.00035728,0.87046
207 | 24,13,0.64897,1.4597,0.3304,0.97406,1,0.75323,0.016021,0.046715,0.012025,0.050266,0.0025203,0.00051116,0.00013299,0.49261
208 | 25,1,0.72162,1.4686,0.43475,0.91337,0.98596,0.60738,0.046324,0.39055,0.097628,0.18471,0.032992,0.0097255,0.001043,1.8296
209 | 25,2,0.71573,1.3808,0.37798,0.95035,0.99123,0.68138,0.025077,0.11445,0.10429,0.23183,0.051004,0.023571,0.00044425,1.5025
210 | 25,3,0.81957,1.56,0.49513,0.91036,0.98421,0.60691,0.045977,0.38473,0.097894,0.20665,0.040954,0.015521,0.0007191,1.5452
211 | 25,4,0.83029,1.633,0.49437,0.92597,0.96842,0.62529,0.045331,0.37399,0.13423,0.2648,0.065523,0.029786,0.00074416,1.7493
212 | 25,5,0.75259,1.5852,0.41155,0.91036,0.99298,0.65821,0.044162,0.35495,0.12959,0.25343,0.060352,0.026003,0.00063907,1.813
213 | 25,6,0.7831,1.5389,0.42639,0.94648,0.97544,0.65478,0.038706,0.27266,0.12607,0.24469,0.056492,0.023259,0.00066766,1.8504
214 | 25,7,0.78823,1.6027,0.44995,0.91357,0.95614,0.62781,0.060707,0.67072,0.12958,0.2522,0.059803,0.025228,0.0009264,1.7489
215 | 25,8,0.83094,1.6643,0.49439,0.96016,0.99649,0.65605,0.023841,0.10345,0.12266,0.24766,0.057789,0.025308,0.00078336,1.7001
216 | 25,9,0.7885,1.638,0.43032,0.95417,1,0.71101,0.026783,0.13055,0.12808,0.21441,0.04395,0.013032,0.0010116,2.2764
217 | 26,1,0.73365,1.5028,0.34887,0.9828,1,0.80144,0.0053647,0.005238,0.029255,0.10327,0.010553,0.0043013,9.1642e-05,0.88407
218 | 26,2,0.82775,1.842,0.48769,0.97383,0.99825,0.68405,0.01619,0.047706,0.022041,0.087963,0.0076781,0.0029651,6.8766e-05,0.68174
219 | 26,3,0.55626,1.2516,0.23975,0.9802,1,0.81411,0.0073087,0.009722,0.056955,0.17933,0.031157,0.017403,8.4502e-05,0.98858
220 | 26,4,0.29901,1.0375,0.10761,0.98386,1,0.84936,0.011462,0.023912,0.032883,0.096034,0.0091383,0.0025695,0.00018734,1.0837
221 | 26,5,0.48223,1.1577,0.16908,0.97886,1,0.71192,0.008864,0.0143,0.037174,0.11384,0.012794,0.0044407,0.00014852,1.0089
222 | 26,6,0.5895,1.2708,0.24364,0.98173,0.99649,0.82082,0.0074388,0.010071,0.041688,0.11576,0.013223,0.0041,0.00025724,1.1105
223 | 26,7,0.322,1.0698,0.13626,0.98017,0.99825,0.81953,0.012913,0.030347,0.076267,0.18383,0.03269,0.014159,0.00026309,1.4955
224 | 26,8,0.482,1.1415,0.15235,0.98536,1,0.85572,0.0072199,0.009487,0.021863,0.079283,0.0062465,0.0017479,0.00013143,0.69178
225 | 26,9,0.87402,2.0754,0.55373,0.96281,0.98421,0.56448,0.019656,0.070316,0.037553,0.10946,0.011839,0.0038499,0.00021407,1.0693
226 | 26,10,0.58138,1.2486,0.21485,0.98704,1,0.83511,0.0046582,0.0039492,0.04072,0.11304,0.012618,0.0036743,0.00033322,1.0368
227 | 26,11,0.40974,1.0732,0.14538,0.98155,1,0.8407,0.0042083,0.0032232,0.04953,0.1489,0.021691,0.010419,0.00010258,1.1781
228 | 26,12,0.82556,1.6903,0.47477,0.97153,1,0.70462,0.019154,0.066774,0.025981,0.10427,0.010756,0.0045965,4.4538e-05,0.67306
229 | 27,1,0.60267,1.254,0.20587,0.98883,1,0.80193,0.0084333,0.012944,0.017777,0.061786,0.003803,0.00091963,0.00015231,0.85839
230 | 27,2,0.66906,1.2751,0.30877,0.9877,1,0.85816,0.0048599,0.0042987,0.031488,0.094241,0.0088032,0.0025337,0.00019057,1.0378
231 | 27,3,0.8138,1.737,0.42956,0.98956,1,0.79117,0.0086131,0.013502,0.033871,0.10399,0.010699,0.0035528,0.00015394,1.0284
232 | 27,4,0.72719,1.4779,0.3298,0.99388,1,0.8423,0.0029668,0.0016019,0.02634,0.081903,0.0066634,0.0017846,0.0001939,0.9805
233 | 27,5,0.79923,1.6549,0.40919,0.98964,1,0.815,0.0048968,0.0043641,0.028693,0.091008,0.0082145,0.0025009,0.00013606,0.98834
234 | 27,6,0.81354,1.6924,0.41016,0.99256,1,0.78909,0.0046348,0.0039096,0.031342,0.096493,0.0092251,0.0028704,0.00017007,1.0171
235 | 27,7,0.8561,1.873,0.49654,0.98047,1,0.6913,0.0092289,0.015501,0.047961,0.1039,0.010681,0.0024623,0.00080703,1.6203
236 | 27,8,0.75968,1.5763,0.38954,0.98731,1,0.82323,0.005582,0.0056708,0.034734,0.098208,0.0095526,0.0027074,0.00023404,1.1433
237 | 27,9,0.8031,1.652,0.40268,0.97695,1,0.79833,0.010883,0.021558,0.046764,0.10227,0.010352,0.0021515,0.00074063,1.5053
238 | 27,10,0.79389,1.5303,0.42937,0.96802,0.99649,0.79209,0.016566,0.049949,0.04703,0.11769,0.013661,0.0041812,0.00034846,1.42
239 | 27,11,0.66551,1.3617,0.26804,0.98949,1,0.76603,0.005179,0.0048816,0.017792,0.074392,0.0055037,0.0018166,5.245e-05,0.62317
240 | 28,1,0.946,3.3475,0.7081,0.96212,1,0.45671,0.035434,0.22851,0.054986,0.13106,0.016886,0.0047646,0.00051238,1.3088
241 | 28,2,0.93038,2.8852,0.67525,0.95738,1,0.50964,0.027111,0.13377,0.050703,0.12329,0.014972,0.0041932,0.00032166,1.3848
242 | 28,3,0.91549,2.6796,0.62862,0.96381,1,0.55874,0.019083,0.066276,0.10351,0.19882,0.038028,0.012751,0.00057453,1.9702
243 | 28,4,0.87202,2.1536,0.53817,0.9788,1,0.65619,0.020327,0.075203,0.081306,0.18726,0.033878,0.013211,0.00036142,1.4606
244 | 28,5,0.91328,2.5654,0.61185,0.98179,1,0.60313,0.006726,0.0082335,0.093484,0.19225,0.035644,0.012245,0.00061603,1.6701
245 | 28,6,0.91996,2.6668,0.63083,0.97894,1,0.58002,0.010311,0.019349,0.1125,0.19785,0.037671,0.011205,0.00078875,2.2241
246 | 28,7,0.91677,2.6955,0.62998,0.98275,1,0.59459,0.0049218,0.0044087,0.10467,0.1876,0.033997,0.0096762,0.00085264,2.0585
247 | 28,8,0.91397,2.7289,0.64656,0.96685,0.99298,0.55519,0.022351,0.090925,0.068514,0.1525,0.022728,0.0070313,0.00041076,1.5681
248 | 28,9,0.8762,2.2436,0.56314,0.98425,1,0.65133,0.010195,0.018915,0.14482,0.22485,0.048122,0.013423,0.0012638,2.4685
249 | 28,10,0.92081,2.9198,0.65845,0.96861,0.99825,0.52299,0.011587,0.024437,0.064788,0.16842,0.027581,0.01122,0.00024357,1.2406
250 | 28,11,0.86832,2.141,0.54004,0.98483,1,0.69313,0.009048,0.0149,0.11927,0.2052,0.040405,0.011659,0.0010933,2.1073
251 | 28,12,0.93323,3.222,0.69174,0.95791,1,0.48142,0.022654,0.093403,0.075307,0.17561,0.029915,0.011062,0.00039957,1.4215
252 | 29,1,0.8375,1.9512,0.4905,0.968,0.98246,0.65138,0.016224,0.047908,0.0051195,0.035621,0.0012673,0.0003224,1.1238e-05,0.23514
253 | 29,2,0.83275,2.0014,0.50157,0.9557,0.99649,0.67019,0.028422,0.14702,0.0092534,0.051082,0.0026026,0.00073282,4.0071e-05,0.32567
254 | 29,3,0.84579,2.0052,0.49931,0.96585,0.99825,0.65149,0.019981,0.07266,0.0050343,0.036189,0.0013079,0.00035119,8.7642e-06,0.22547
255 | 29,4,0.74053,1.6378,0.3892,0.9666,1,0.74844,0.025032,0.11404,0.0083497,0.045391,0.0020561,0.00054932,2.9454e-05,0.38449
256 | 29,5,0.82924,1.8903,0.47248,0.96435,0.99825,0.661,0.023836,0.10341,0.0057361,0.037634,0.0014143,0.00036897,1.3273e-05,0.27099
257 | 29,6,0.82256,1.9016,0.47622,0.96505,1,0.69746,0.025026,0.11398,0.0056632,0.033415,0.0011153,0.00022942,3.7442e-05,0.29502
258 | 29,7,0.84501,2.0918,0.51732,0.9451,1,0.6071,0.039781,0.28802,0.0081002,0.044849,0.0020074,0.00050789,2.9215e-05,0.34316
259 | 29,8,0.8187,1.9126,0.47835,0.94904,0.99298,0.58585,0.03806,0.26364,0.0050219,0.034026,0.0011564,0.00029213,1.5463e-05,0.26037
260 | 29,9,0.78965,1.8186,0.45353,0.94929,1,0.63343,0.030371,0.16787,0.008049,0.04345,0.0018843,0.00045429,2.9147e-05,0.36533
261 | 29,10,0.80728,1.857,0.46197,0.95012,0.99298,0.60913,0.028443,0.14724,0.0096387,0.060095,0.0035984,0.0013452,1.6802e-05,0.27688
262 | 29,11,0.80393,1.8829,0.4698,0.93066,0.99298,0.58011,0.043787,0.34894,0.010183,0.061975,0.0038262,0.0014329,1.8914e-05,0.28509
263 | 29,12,0.85451,2.06,0.51569,0.968,0.98421,0.60909,0.025453,0.11791,0.0079224,0.044948,0.0020163,0.00051981,2.6189e-05,0.32598
264 | 30,1,0.50924,1.2166,0.27943,0.93462,0.99123,0.4092,0.016089,0.047113,0.096518,0.17131,0.028511,0.006963,0.0015453,1.851
265 | 30,2,0.52914,1.2066,0.25595,0.93945,0.97895,0.51693,0.019009,0.065765,0.14675,0.19841,0.037876,0.0073471,0.0023476,2.7085
266 | 30,3,0.2458,1.1011,0.20617,0.94995,0.99298,0.53189,0.012818,0.029901,0.14279,0.20537,0.04047,0.0090273,0.0018094,2.579
267 | 30,4,0.55518,1.2318,0.25226,0.94783,0.99123,0.54965,0.015115,0.041582,0.059148,0.13996,0.019211,0.0057336,0.00042099,1.3579
268 | 30,5,0.35141,1.1428,0.24265,0.93638,0.99123,0.47846,0.019721,0.070781,0.1074,0.18794,0.034115,0.0089893,0.0015026,1.9019
269 | 30,6,0.4803,1.2022,0.26797,0.94544,0.99825,0.52139,0.022778,0.09443,0.066259,0.14818,0.021485,0.0061942,0.00073522,1.3669
270 | 30,7,0.19287,1.0551,0.25044,0.93641,0.99474,0.47284,0.019693,0.070581,0.055935,0.1303,0.016694,0.0045832,0.00044001,1.4111
271 | 30,8,0.49634,1.1832,0.22855,0.94687,0.98947,0.58613,0.017327,0.054638,0.047106,0.12748,0.015992,0.0051731,0.00023865,1.1489
272 | 30,9,0.33254,1.1208,0.27473,0.93625,0.99474,0.49836,0.024394,0.1083,0.10809,0.16882,0.027709,0.0059808,0.0012335,2.4866
273 | 30,10,0.39606,1.1647,0.29415,0.94064,0.99298,0.5486,0.025244,0.11598,0.051625,0.12014,0.014228,0.003721,0.00038216,1.4943
274 | 30,11,0.41933,1.1928,0.30567,0.93428,0.98947,0.50688,0.020948,0.079866,0.057408,0.12508,0.015404,0.003903,0.00045984,1.6497
275 | 30,12,0.45767,1.1629,0.24078,0.94275,0.99298,0.63919,0.020279,0.074844,0.028432,0.095746,0.0090841,0.002812,0.0001478,0.78535
276 | 31,1,0.99541,9.4912,0.91397,0.84991,0.97368,0.17429,0.082108,1.227,0.012512,0.060688,0.0036695,0.0010735,5.216e-05,0.41928
277 | 31,2,0.99611,11.006,0.92239,0.86962,0.99123,0.14581,0.064792,0.76404,0.011176,0.057808,0.0033307,0.00098398,3.9298e-05,0.3731
278 | 31,3,0.99767,12.009,0.94529,0.67005,0.94386,0.11114,0.19898,7.2062,0.010215,0.051104,0.0026048,0.00066765,4.5324e-05,0.39905
279 | 31,4,0.99456,9.7351,0.90444,0.93346,0.98421,0.19638,0.020897,0.079479,0.016599,0.075247,0.0056302,0.0019012,4.441e-05,0.48992
280 | 31,5,0.997,12.548,0.92397,0.93876,0.96842,0.16105,0.026941,0.1321,0.010116,0.057812,0.0033311,0.0010891,2.9179e-05,0.3101
281 | 31,6,0.99698,10.913,0.93836,0.70085,0.93684,0.11552,0.15334,4.2796,0.010649,0.055629,0.0030851,0.00089618,3.6128e-05,0.37415
282 | 31,7,0.99551,9.6915,0.91233,0.88256,0.93509,0.17567,0.049695,0.44946,0.013721,0.067649,0.0045556,0.0015063,4.4371e-05,0.41054
283 | 31,8,0.99502,8.991,0.92789,0.68687,0.86842,0.14478,0.18853,6.4689,0.013268,0.068973,0.0047347,0.001673,3.4802e-05,0.37492
284 | 31,9,0.99583,9.1366,0.92503,0.69234,0.90702,0.15101,0.17918,5.8432,0.0054072,0.044172,0.0019474,0.00070285,6.9241e-06,0.1694
285 | 31,10,0.99593,10.12,0.92461,0.80662,0.90702,0.15053,0.12257,2.7342,0.0082902,0.052333,0.0027313,0.00090459,1.5362e-05,0.27303
286 | 31,11,0.99422,9.1985,0.91119,0.8375,0.98421,0.17712,0.055414,0.55887,0.016168,0.079996,0.0063587,0.0024534,5.5348e-05,0.38686
287 | 32,1,0.86224,2.0735,0.52269,0.98686,0.99474,0.70529,0.010097,0.018554,0.041404,0.12163,0.014579,0.0048689,0.00027608,0.9458
288 | 32,2,0.87024,2.1094,0.52863,0.9836,0.99298,0.60784,0.0031737,0.0018332,0.026902,0.091391,0.0082832,0.0024385,0.00016093,0.73904
289 | 32,3,0.86248,2.0398,0.51338,0.98831,1,0.71419,0.0057488,0.0060148,0.052681,0.13941,0.019064,0.006376,0.00045928,1.0667
290 | 32,4,0.90305,2.3455,0.57721,0.98504,1,0.58743,0.0044044,0.0035306,0.022443,0.077826,0.0060204,0.0015394,0.00021043,0.68561
291 | 32,5,0.88566,2.2029,0.54993,0.98635,1,0.59039,0.0037601,0.0025732,0.039469,0.12459,0.015285,0.0056604,0.00020409,0.84271
292 | 32,6,0.88078,2.147,0.54028,0.98989,1,0.71926,0.0031646,0.0018226,0.044894,0.12906,0.016385,0.0056383,0.00032291,0.96576
293 | 32,7,0.88485,2.2398,0.55754,0.97997,0.99825,0.67974,0.0091286,0.015166,0.025658,0.087206,0.0075476,0.0021518,0.00017873,0.75154
294 | 32,8,0.88661,2.2553,0.56472,0.98284,1,0.67537,0.012123,0.026749,0.022251,0.083617,0.0069433,0.0020893,0.00016304,0.60779
295 | 32,9,0.83188,1.8445,0.46807,0.98691,1,0.75011,0.0049353,0.0044331,0.034814,0.10867,0.011672,0.0036787,0.0003005,0.80348
296 | 32,10,0.85278,1.9935,0.50107,0.98449,0.99825,0.72127,0.0068277,0.0084843,0.015559,0.070923,0.005005,0.0015733,6.9367e-05,0.44798
297 | 32,11,0.78676,1.6815,0.41383,0.97686,0.99825,0.7686,0.013524,0.033289,0.016538,0.076334,0.0057932,0.0019786,7.0661e-05,0.43227
298 | 33,1,0.69887,1.4947,0.34694,0.97906,1,0.74423,0.0070445,0.0090318,0.067277,0.15809,0.024383,0.0079843,0.00066857,1.2554
299 | 33,2,0.60555,1.3891,0.29569,0.97439,1,0.73603,0.010821,0.02131,0.1363,0.20829,0.04158,0.010017,0.0020084,2.2733
300 | 33,3,0.79495,1.7056,0.42041,0.98477,1,0.76317,0.0074486,0.010098,0.064127,0.14621,0.020929,0.0061549,0.00065287,1.3662
301 | 33,4,0.75535,1.6662,0.4065,0.97627,1,0.75126,0.0098445,0.017638,0.020477,0.09052,0.0081274,0.0031335,0.00012137,0.4391
302 | 33,5,0.69398,1.4542,0.32457,0.98263,0.99474,0.79449,0.0063863,0.0074227,0.079016,0.16552,0.026666,0.0079652,0.00077178,1.5438
303 | 33,6,0.81992,1.8698,0.47287,0.97829,0.99825,0.65353,0.0074141,0.010004,0.06789,0.142,0.019765,0.004991,0.00096699,1.4857
304 | 33,7,0.68878,1.4399,0.32986,0.98318,1,0.7847,0.0070802,0.0091235,0.083675,0.17243,0.028872,0.008786,0.0010143,1.5313
305 | 33,8,0.72174,1.5301,0.35707,0.97994,0.99649,0.78622,0.013545,0.03339,0.040677,0.11435,0.012908,0.0038198,0.00036093,0.98926
306 | 33,9,0.73067,1.5145,0.34921,0.98686,1,0.81119,0.017693,0.056975,0.081485,0.16841,0.02758,0.0081927,0.00084762,1.552
307 | 33,10,0.65119,1.4026,0.30387,0.9799,1,0.80081,0.0082573,0.012409,0.10567,0.185,0.033093,0.0087245,0.0013911,1.9632
308 | 33,11,0.71999,1.569,0.37089,0.97599,0.99825,0.77145,0.013975,0.035546,0.11997,0.19286,0.035863,0.0086099,0.001837,2.1311
309 | 34,1,0.99571,10.352,0.90609,0.90652,0.97895,0.16732,0.025787,0.12103,0.023523,0.089455,0.0079387,0.0028882,6.9318e-05,0.73774
310 | 34,2,0.99569,9.9653,0.90578,0.91515,0.9807,0.17113,0.023896,0.10392,0.020381,0.082596,0.0067759,0.0024074,5.0484e-05,0.67264
311 | 34,3,0.9967,11.361,0.914,0.91815,0.93684,0.14435,0.019976,0.072629,0.020667,0.085514,0.0072595,0.0028469,4.6487e-05,0.67925
312 | 34,4,0.99841,16.832,0.94116,0.85531,0.9,0.094537,0.035845,0.23384,0.012341,0.066682,0.0044268,0.0017052,1.7724e-05,0.41828
313 | 34,5,0.99799,15.068,0.93667,0.8807,0.95789,0.12132,0.032164,0.18828,0.011686,0.054827,0.002997,0.00086356,4.3787e-05,0.50576
314 | 34,6,0.99512,10.377,0.90564,0.92135,0.99825,0.17941,0.016647,0.050433,0.0204,0.071662,0.0051092,0.0012661,0.00017021,0.71514
315 | 34,7,0.99505,10.736,0.90851,0.92586,0.99649,0.18236,0.028571,0.14856,0.017761,0.06961,0.0048222,0.0013129,0.00013111,0.59218
316 | 34,8,0.99518,10.421,0.90385,0.96551,0.98947,0.20051,0.0077228,0.010855,0.021258,0.080385,0.0064203,0.0019168,9.5565e-05,0.66405
317 | 34,9,0.9953,10.63,0.90598,0.88866,0.95789,0.15967,0.025636,0.11961,0.014782,0.065416,0.0042611,0.0012831,6.4365e-05,0.52178
318 | 34,10,0.99515,10.007,0.90328,0.91594,0.97719,0.16244,0.033129,0.19975,0.017258,0.070646,0.004966,0.0014641,7.1555e-05,0.59946
319 | 34,11,0.99871,19.038,0.94834,0.851,0.90702,0.086183,0.073048,0.97117,0.0078165,0.048089,0.0023072,0.00075332,1.2922e-05,0.34029
320 | 35,1,0.9109,2.5488,0.6106,0.97388,0.99825,0.55818,0.01972,0.070775,0.10132,0.17022,0.028158,0.0074614,0.0010974,2.3847
321 | 35,2,0.90391,2.458,0.5977,0.9861,1,0.61429,0.0067013,0.0081731,0.08902,0.16706,0.027151,0.007219,0.00096593,1.8937
322 | 35,3,0.90755,2.582,0.62394,0.96837,0.99825,0.55674,0.031714,0.18305,0.079387,0.16213,0.025613,0.0074124,0.00069878,1.6951
323 | 35,4,0.91708,2.6498,0.62919,0.98493,1,0.57926,0.0068053,0.0084289,0.082187,0.16877,0.027693,0.0085027,0.00064002,1.7157
324 | 35,5,0.9166,2.6711,0.6333,0.98228,1,0.51619,0.017465,0.055514,0.10368,0.181,0.031721,0.0081985,0.0014211,1.973
325 | 35,6,0.8942,2.4203,0.59323,0.98407,1,0.60794,0.01743,0.055294,0.10102,0.18748,0.033957,0.0098155,0.0011793,1.8275
326 | 35,7,0.93847,3.0198,0.67312,0.98733,1,0.53234,0.0076393,0.010621,0.07629,0.18017,0.03144,0.011932,0.00072251,1.2434
327 | 35,8,0.89518,2.3875,0.581,0.98796,1,0.63833,0.0077926,0.011052,0.07304,0.1591,0.024688,0.0073807,0.00084138,1.408
328 | 35,9,0.91707,2.6504,0.63359,0.96002,0.99298,0.53972,0.012062,0.026481,0.10121,0.18433,0.032861,0.0089657,0.0014149,1.7979
329 | 35,10,0.92535,2.803,0.65133,0.976,1,0.48413,0.015435,0.04336,0.098946,0.18338,0.032533,0.0091801,0.0013297,1.8033
330 | 35,11,0.91861,2.8114,0.64707,0.94843,0.95614,0.51186,0.045037,0.36915,0.041345,0.13297,0.017374,0.0070233,0.00018923,0.83183
331 | 36,1,0.39093,1.1025,0.73351,0.72022,0.69474,0.17954,0.076072,1.0532,0.059213,0.15747,0.024197,0.0095408,0.00024731,1.2042
332 | 36,2,0.47124,1.1349,0.81159,0.65915,0.47368,0.093982,0.096492,1.6945,0.098618,0.21062,0.042478,0.016848,0.00058101,1.5913
333 | 36,3,0.3687,1.0456,0.77124,0.74413,0.77368,0.22278,0.075187,1.0289,0.074488,0.17993,0.031359,0.012414,0.0003469,1.3541
334 | 36,4,0.14986,1.0558,0.77733,0.73454,0.66316,0.1688,0.08041,1.1768,0.10395,0.1806,0.031585,0.0080165,0.0014315,1.9975
335 | 36,5,0.68069,1.1866,0.78745,0.73496,0.6,0.14029,0.072447,0.95524,0.09277,0.18451,0.032923,0.010852,0.00059383,1.8568
336 | 36,6,0.37522,1.1417,0.81725,0.68511,0.58772,0.12523,0.09186,1.5358,0.11488,0.20861,0.041703,0.013344,0.00082033,2.0281
337 | 36,7,0.28064,1.0849,0.75319,0.72152,0.71404,0.13686,0.078996,1.1358,0.14122,0.2183,0.045488,0.012002,0.0015154,2.4059
338 | 36,8,0.35344,1.0329,0.78147,0.70737,0.61579,0.13503,0.089763,1.4664,0.097663,0.20703,0.041101,0.016123,0.00045288,1.6935
339 | 36,9,0.59988,1.1427,0.71532,0.66101,0.47544,0.15747,0.11337,2.3394,0.050389,0.13585,0.018121,0.00619,0.00026454,1.1526
340 | 36,10,0.47195,1.0901,0.85409,0.53598,0.39649,0.078376,0.13227,3.184,0.082007,0.18782,0.034074,0.013487,0.00032855,1.5623
341 | 


--------------------------------------------------------------------------------
/examples/leafDataset/logreg.js:
--------------------------------------------------------------------------------
 1 | const LogisticRegression = require('ml-logistic-regression');
 2 | const csv = require('csvtojson');
 3 | const {Matrix} = require('ml-matrix');
 4 | const PCA = require('ml-pca');
 5 | 
 6 | 
 7 | let logreg = new LogisticRegression({numSteps: 5000, learningRate: 5e-3});
 8 | 
 9 | const csvFilePath = 'leaf.csv'; // Data
10 | const names = ['type', 'specimenNumber', 'eccentricity', 'aspectRatio', 'elongation', 'solidity', 'stochasticConvexity', 'isoperimetricFactor', 'maxIndetationDepth', 'lobedness', 'intensity', 'contrast', 'smoothness', 'thirdMoment', 'uniformity', 'entropy']; // For header
11 | 
12 | let seperationSize;
13 | let data = [], X = [], y = [];
14 | let trainingSetX = [], trainingSetY = [], testSetX = [], testSetY = [];
15 | 
16 | csv({noheader: true, headers: names})
17 |     .fromFile(csvFilePath)
18 |     .on('json', (jsonObj) => {
19 |         data.push(jsonObj); // Push each object to data Array
20 |     })
21 |     .on('done', (error) => {
22 |         seperationSize = 0.9 * data.length;
23 |         data = shuffleArray(data);
24 |         dressData();
25 |     });
26 | 
27 | function dressData() {
28 |     let types = new Set(); // To gather UNIQUE classes
29 |     data.forEach((row) => {
30 |         types.add(row.type);
31 |     });
32 |     typesArray = [...types]; // To save the different types of classes.
33 | 
34 |     data.forEach((row) => {
35 |         let rowArray, typeNumber;
36 |         rowArray = Object.keys(row).map(key => parseFloat(row[key])).slice(2, 16); // We don't use the 2 first elements, which are the type (i.e class) and the specimen number (i.e ID)
37 |         typeNumber = typesArray.indexOf(row.type); // Convert type(String) to type(Number)
38 | 
39 |         X.push(rowArray);
40 |         y.push(typeNumber);
41 |     });
42 | 
43 |     trainingSetX = new Matrix(X.slice(0, seperationSize));
44 |     trainingSetY = Matrix.columnVector(y.slice(0, seperationSize));
45 |     testSetX = new Matrix(X.slice(seperationSize));
46 |     testSetY = y.slice(seperationSize);
47 |     train();
48 | }
49 | 
50 | function train() {
51 |     logreg.train(trainingSetX, trainingSetY);
52 |     test();
53 | }
54 | 
55 | function test() {
56 |     const result = logreg.predict(new Matrix(testSetX));
57 |     const testSetLength = testSetX.length
58 |     const predictionError = error(result, testSetY);
59 |     console.log(`Test Set Size = ${testSetLength} and number of Misclassifications = ${predictionError}`);
60 | }
61 | 
62 | function error(predicted, expected) {
63 |     let misclassifications = 0;
64 |     for (var index = 0; index < predicted.length; index++) {
65 |         console.log(`truth : ${expected[index]} and prediction : ${predicted[index]}`);
66 |         if (predicted[index] !== expected[index]) {
67 |             misclassifications++;
68 |         }
69 |     }
70 |     return misclassifications;
71 | }
72 | 
73 | function shuffleArray(array) {
74 |     for (var i = array.length - 1; i > 0; i--) {
75 |         var j = Math.floor(Math.random() * (i + 1));
76 |         var temp = array[i];
77 |         array[i] = array[j];
78 |         array[j] = temp;
79 |     }
80 |     return array;
81 | }
82 | 


--------------------------------------------------------------------------------
/examples/leafDataset/naive-bayes.js:
--------------------------------------------------------------------------------
 1 | const Bayes = require('ml-naivebayes');
 2 | const csv = require('csvtojson');
 3 | 
 4 | 
 5 | bayes = new Bayes();
 6 | 
 7 | const csvFilePath = 'leaf.csv'; // Data
 8 | const names = ['type', 'specimenNumber', 'eccentricity', 'aspectRatio', 'elongation', 'solidity', 'stochasticConvexity', 'isoperimetricFactor', 'maxIndetationDepth', 'lobedness', 'intensity', 'contrast', 'smoothness', 'thirdMoment', 'uniformity', 'entropy']; // For header
 9 | 
10 | let seperationSize;
11 | let data = [], X = [], y = [];
12 | let trainingSetX = [], trainingSetY = [], testSetX = [], testSetY = [];
13 | 
14 | csv({noheader: true, headers: names})
15 |     .fromFile(csvFilePath)
16 |     .on('json', (jsonObj) => {
17 |         data.push(jsonObj); // Push each object to data Array
18 |     })
19 |     .on('done', (error) => {
20 |         seperationSize = 0.9 * data.length;
21 |         data = shuffleArray(data);
22 |         dressData();
23 |     });
24 | 
25 | function dressData() {
26 |     let types = new Set(); // To gather UNIQUE classes
27 |     data.forEach((row) => {
28 |         types.add(row.type);
29 |     });
30 |     typesArray = [...types]; // To save the different types of classes.
31 | 
32 |     data.forEach((row) => {
33 |         let rowArray, typeNumber;
34 |         rowArray = Object.keys(row).map(key => parseFloat(row[key])).slice(2, 16); // We don't use the 2 first elements, which are the type (i.e class) and the specimen number (i.e ID)
35 |         typeNumber = typesArray.indexOf(row.type); // Convert type(String) to type(Number)
36 | 
37 |         X.push(rowArray);
38 |         y.push(typeNumber);
39 |     });
40 | 
41 |     trainingSetX = X.slice(0, seperationSize);
42 |     trainingSetY = y.slice(0, seperationSize);
43 |     testSetX = X.slice(seperationSize);
44 |     testSetY = y.slice(seperationSize);
45 | 
46 |     train();
47 | }
48 | 
49 | function train() {
50 |     bayes.train(trainingSetX, trainingSetY);
51 |     test();
52 | }
53 | 
54 | function test() {
55 |     const result = bayes.predict(testSetX);
56 |     const testSetLength = testSetX.length
57 |     const predictionError = error(result, testSetY);
58 |     console.log(`Test Set Size = ${testSetLength} and number of Misclassifications = ${predictionError}`);
59 | }
60 | 
61 | function error(predicted, expected) {
62 |     let misclassifications = 0;
63 |     for (var index = 0; index < predicted.length; index++) {
64 |         console.log(`truth : ${expected[index]} and prediction : ${predicted[index]}`);
65 |         if (predicted[index] !== expected[index]) {
66 |             misclassifications++;
67 |         }
68 |     }
69 |     return misclassifications;
70 | }
71 | 
72 | function shuffleArray(array) {
73 |     for (var i = array.length - 1; i > 0; i--) {
74 |         var j = Math.floor(Math.random() * (i + 1));
75 |         var temp = array[i];
76 |         array[i] = array[j];
77 |         array[j] = temp;
78 |     }
79 |     return array;
80 | }
81 | 


--------------------------------------------------------------------------------
/examples/leafDataset/package.json:
--------------------------------------------------------------------------------
 1 | {
 2 |   "name": "examples of classification",
 3 |   "version": "1.0.0",
 4 |   "description": "use methods of mljs on a dataset from UCI",
 5 |   "keywords": [
 6 |     "SVM",
 7 |     "KNN",
 8 |     "MLJS",
 9 |     "classification",
10 |     "UCI dataset"
11 |   ],
12 |   "author": "jajoe",
13 |   "license": "MIT",
14 |   "dependencies": {
15 |     "csvtojson": "^1.1.7",
16 |     "libsvm-js": "^0.1.1",
17 |     "ml-knn": "^2.1.0",
18 |     "ml-logistic-regression": "^1.0.2",
19 |     "ml-matrix": "^5.0.0",
20 |     "ml-naivebayes": "^2.0.0",
21 |     "ml-pca": "^2.1.0"
22 |   }
23 | }
24 | 


--------------------------------------------------------------------------------
/examples/leafDataset/svm.js:
--------------------------------------------------------------------------------
 1 | const SVM = require('libsvm-js/asm');
 2 | const csv = require('csvtojson');
 3 | 
 4 | let options = {
 5 |     kernel : SVM.KERNEL_TYPES.POLYNOMIAL,
 6 |     degree : 3
 7 | }
 8 | 
 9 | const csvFilePath = 'leaf.csv'; // Data
10 | const names = ['type', 'specimenNumber', 'eccentricity', 'aspectRatio', 'elongation', 'solidity', 'stochasticConvexity', 'isoperimetricFactor', 'maxIndetationDepth', 'lobedness', 'intensity', 'contrast', 'smoothness', 'thirdMoment', 'uniformity', 'entropy']; // For header
11 | 
12 | svm = new SVM(options);
13 | 
14 | let seperationSize;
15 | let data = [], X = [], y = [];
16 | let trainingSetX = [], trainingSetY = [], testSetX = [], testSetY = [];
17 | 
18 | csv({noheader: true, headers: names})
19 |     .fromFile(csvFilePath)
20 |     .on('json', (jsonObj) => {
21 |         data.push(jsonObj); // Push each object to data Array
22 |     })
23 |     .on('done', (error) => {
24 |         seperationSize = 0.9 * data.length;
25 |         data = shuffleArray(data);
26 |         dressData();
27 |     });
28 | 
29 | function dressData() {
30 |     let types = new Set(); // To gather UNIQUE classes
31 |     data.forEach((row) => {
32 |         types.add(row.type);
33 |     });
34 |     typesArray = [...types]; // To save the different types of classes.
35 | 
36 |     data.forEach((row) => {
37 |         let rowArray, typeNumber;
38 |         rowArray = Object.keys(row).map(key => parseFloat(row[key])).slice(2, 16); // We don't use the 2 first elements, which are the type (i.e class) and the specimen number (i.e ID)
39 |         typeNumber = typesArray.indexOf(row.type); // Convert type(String) to type(Number)
40 | 
41 |         X.push(rowArray);
42 |         y.push(typeNumber);
43 |     });
44 | 
45 |     trainingSetX = X.slice(0, seperationSize);
46 |     trainingSetY = y.slice(0, seperationSize);
47 |     testSetX = X.slice(seperationSize);
48 |     testSetY = y.slice(seperationSize);
49 | 
50 |     train();
51 | }
52 | 
53 | function train() {
54 |     svm.train(trainingSetX, trainingSetY);
55 |     test();
56 | }
57 | 
58 | function test() {
59 |     const result = svm.predict(testSetX);
60 |     const testSetLength = testSetX.length
61 |     const predictionError = error(result, testSetY);
62 |     console.log(`Test Set Size = ${testSetLength} and number of Misclassifications = ${predictionError}`);
63 | }
64 | 
65 | function error(predicted, expected) {
66 |     let misclassifications = 0;
67 |     for (var index = 0; index < predicted.length; index++) {
68 |         console.log(`truth : ${expected[index]} and prediction : ${predicted[index]}`);
69 |         if (predicted[index] !== expected[index]) {
70 |             misclassifications++;
71 |         }
72 |     }
73 |     return misclassifications;
74 | }
75 | 
76 | 
77 | function shuffleArray(array) {
78 |     for (var i = array.length - 1; i > 0; i--) {
79 |         var j = Math.floor(Math.random() * (i + 1));
80 |         var temp = array[i];
81 |         array[i] = array[j];
82 |         array[j] = temp;
83 |     }
84 |     return array;
85 | }
86 | 


--------------------------------------------------------------------------------
/examples/sg.js:
--------------------------------------------------------------------------------
1 | const savitzkyGolay = require('../dist/ml.js').savitzkyGolay;
2 | 
3 | let result = savitzkyGolay([1, 2, 3, 4, 5, 4, 3, 2, 1], 1, { derivative: 0 });
4 | 
5 | console.log(result);
6 | 


--------------------------------------------------------------------------------
/package.json:
--------------------------------------------------------------------------------
  1 | {
  2 |   "name": "ml",
  3 |   "version": "8.0.0",
  4 |   "description": "Machine learning tools",
  5 |   "main": "src/index.js",
  6 |   "scripts": {
  7 |     "build": "cheminfo-build --root ML",
  8 |     "compile": "rollup -c",
  9 |     "eslint": "eslint src --cache",
 10 |     "eslint-fix": "npm run eslint -- --fix",
 11 |     "prepack": "npm run compile",
 12 |     "prettier": "prettier --check src",
 13 |     "prettier-write": "prettier --write src",
 14 |     "test": "npm run test-only && npm run eslint && npm run prettier",
 15 |     "test-only": "vitest run --coverage"
 16 |   },
 17 |   "files": [
 18 |     "src",
 19 |     "dist",
 20 |     "lib"
 21 |   ],
 22 |   "repository": {
 23 |     "type": "git",
 24 |     "url": "https://github.com/mljs/ml.git"
 25 |   },
 26 |   "keywords": [
 27 |     "machine",
 28 |     "learning",
 29 |     "data",
 30 |     "mining",
 31 |     "datamining"
 32 |   ],
 33 |   "author": "Michaël Zasso",
 34 |   "contributors": [
 35 |     "Miguel Asencio <maasencioh@gmail.com> (https://github.com/maasencioh)",
 36 |     "Jefferson Hernández"
 37 |   ],
 38 |   "license": "MIT",
 39 |   "bugs": {
 40 |     "url": "https://github.com/mljs/ml/issues"
 41 |   },
 42 |   "homepage": "https://github.com/mljs/ml",
 43 |   "dependencies": {
 44 |     "binary-search": "^1.3.6",
 45 |     "ml-array-max": "^1.2.4",
 46 |     "ml-array-mean": "^1.1.6",
 47 |     "ml-array-median": "^1.1.6",
 48 |     "ml-array-min": "^1.2.3",
 49 |     "ml-array-mode": "^1.1.5",
 50 |     "ml-array-normed": "^1.3.7",
 51 |     "ml-array-rescale": "^1.3.7",
 52 |     "ml-array-sequential-fill": "^1.1.8",
 53 |     "ml-array-standard-deviation": "^1.1.8",
 54 |     "ml-array-sum": "^1.1.6",
 55 |     "ml-array-variance": "^1.1.8",
 56 |     "ml-array-xy-centroids-merge": "^1.0.2",
 57 |     "ml-array-xy-covariance": "^0.2.1",
 58 |     "ml-array-xy-equally-spaced": "^1.2.1",
 59 |     "ml-array-xy-filter-x": "1.0.2",
 60 |     "ml-array-xy-max-merge": "^1.0.2",
 61 |     "ml-array-xy-max-y": "^1.0.2",
 62 |     "ml-array-xy-sort-x": "^1.0.2",
 63 |     "ml-array-xy-weighted-merge": "^1.0.2",
 64 |     "ml-arrayxy-closestx": "^2.0.2",
 65 |     "ml-arrayxy-uniquex": "^1.0.2",
 66 |     "ml-bit-array": "^1.0.0",
 67 |     "ml-cart": "^2.1.1",
 68 |     "ml-confusion-matrix": "^2.0.0",
 69 |     "ml-cross-validation": "^1.3.0",
 70 |     "ml-distance": "^4.0.1",
 71 |     "ml-distance-matrix": "^2.0.1",
 72 |     "ml-fcnnls": "^3.0.0",
 73 |     "ml-fnn": "^5.0.0",
 74 |     "ml-gsd": "^12.1.8",
 75 |     "ml-hash-table": "^1.0.0",
 76 |     "ml-hclust": "^3.1.0",
 77 |     "ml-kernel": "^3.0.0",
 78 |     "ml-kmeans": "^6.0.0",
 79 |     "ml-knn": "^3.0.0",
 80 |     "ml-levenberg-marquardt": "^4.1.3",
 81 |     "ml-matrix": "^6.12.0",
 82 |     "ml-naivebayes": "^4.0.0",
 83 |     "ml-ngmca": "^1.0.0",
 84 |     "ml-pad-array": "^2.0.0",
 85 |     "ml-pca": "^4.1.1",
 86 |     "ml-performance": "^0.2.0",
 87 |     "ml-pls": "^4.3.2",
 88 |     "ml-random": "^2.0.0",
 89 |     "ml-random-forest": "^2.1.0",
 90 |     "ml-regression": "^6.2.0",
 91 |     "ml-savitzky-golay": "^5.0.0",
 92 |     "ml-som": "^0.0.6",
 93 |     "ml-sparse-matrix": "^2.1.0",
 94 |     "ml-xsadd": "^3.0.1",
 95 |     "num-sort": "^3.0.0"
 96 |   },
 97 |   "devDependencies": {
 98 |     "@babel/plugin-transform-modules-commonjs": "^7.25.7",
 99 |     "@vitest/coverage-v8": "^2.1.3",
100 |     "cheminfo-build": "^1.2.0",
101 |     "eslint": "^9.12.0",
102 |     "eslint-config-cheminfo": "^12.0.1",
103 |     "esm": "^3.2.25",
104 |     "prettier": "^3.3.3",
105 |     "rollup": "^4.24.0",
106 |     "vitest": "^2.1.3"
107 |   }
108 | }
109 | 


--------------------------------------------------------------------------------
/rollup.config.mjs:
--------------------------------------------------------------------------------
1 | export default {
2 |   input: 'src/index.js',
3 |   output: {
4 |     format: 'cjs',
5 |     file: 'lib/index.js'
6 |   }
7 | };
8 | 


--------------------------------------------------------------------------------
/src/__tests__/index.test.js:
--------------------------------------------------------------------------------
 1 | import { it, describe, expect } from 'vitest';
 2 | 
 3 | import * as ML from '..';
 4 | 
 5 | describe('Check all properties are defined', () => {
 6 |   const objectsAndNbKeys = {
 7 |     Distance: 42,
 8 |     Similarity: 11,
 9 |     Array: 11,
10 |     ArrayXY: 10,
11 |     BitArray: 12,
12 |     KMeans: 2,
13 |     HClust: 2,
14 |     NaiveBayes: 2,
15 |     CrossValidation: 6,
16 |     FCNNLS: 2,
17 |     MatrixLib: 36,
18 |     GSD: 7,
19 |   };
20 | 
21 |   it.each(Object.keys(ML))('Check %s', (key) => {
22 |     // if a function it should key should start with lowercase
23 |     // if a class it should key should start with uppercase
24 |     expect(ML[key]).toBeDefined();
25 | 
26 |     if (objectsAndNbKeys[key]) {
27 |       expect(typeof ML[key]).toBe('object');
28 |       expect(Object.keys(ML[key]).length).toBe(objectsAndNbKeys[key]);
29 |     } else if (key[0] === key[0].toUpperCase()) {
30 |       // should be a class
31 |       expect(typeof ML[key]).toBe('function');
32 |       expect(typeof ML[key].prototype).toBe('object');
33 |       expect(ML[key].prototype?.constructor).toBe(ML[key]);
34 |     } else {
35 |       // should be function
36 |       expect(typeof ML[key]).toBe('function');
37 |     }
38 |   });
39 | });
40 | 


--------------------------------------------------------------------------------
/src/index.js:
--------------------------------------------------------------------------------
  1 | /* eslint-disable import/newline-after-import */
  2 | /* eslint-disable import/order */
  3 | /* eslint-disable import/first */
  4 | 
  5 | // Try to keep this list in the same structure as the README.
  6 | 
  7 | // Unsupervised learning
  8 | export { PCA } from 'ml-pca';
  9 | 
 10 | // Supervised learning
 11 | 
 12 | export { default as KNN } from 'ml-knn';
 13 | export { PLS, KOPLS, OPLS, oplsNipals } from 'ml-pls';
 14 | 
 15 | export { ConfusionMatrix } from 'ml-confusion-matrix';
 16 | 
 17 | // Artificial neural networks
 18 | export { default as FNN } from 'ml-fnn';
 19 | export { default as SOM } from 'ml-som';
 20 | 
 21 | // Regression
 22 | export {
 23 |   SimpleLinearRegression,
 24 |   PolynomialRegression,
 25 |   MultivariateLinearRegression,
 26 |   PowerRegression,
 27 |   ExponentialRegression,
 28 |   TheilSenRegression,
 29 |   RobustPolynomialRegression,
 30 | } from 'ml-regression';
 31 | 
 32 | // Optimization
 33 | export { levenbergMarquardt } from 'ml-levenberg-marquardt';
 34 | 
 35 | // Math
 36 | import * as MatrixLib from 'ml-matrix';
 37 | const {
 38 |   Matrix,
 39 |   SVD,
 40 |   EVD,
 41 |   CholeskyDecomposition,
 42 |   LuDecomposition,
 43 |   QrDecomposition,
 44 | } = MatrixLib;
 45 | export {
 46 |   Matrix,
 47 |   SVD,
 48 |   EVD,
 49 |   CholeskyDecomposition,
 50 |   LuDecomposition,
 51 |   QrDecomposition,
 52 | };
 53 | 
 54 | export { SparseMatrix } from 'ml-sparse-matrix';
 55 | export { default as Kernel } from 'ml-kernel';
 56 | export { distance as Distance, similarity as Similarity } from 'ml-distance';
 57 | export { default as distanceMatrix } from 'ml-distance-matrix';
 58 | export { XSadd } from 'ml-xsadd';
 59 | export { nGMCA } from 'ml-ngmca';
 60 | 
 61 | // Statistics
 62 | export { default as Performance } from 'ml-performance';
 63 | 
 64 | // Data preprocessing
 65 | export { default as savitzkyGolay } from 'ml-savitzky-golay';
 66 | 
 67 | // Utility
 68 | export { default as BitArray } from 'ml-bit-array';
 69 | export { default as HashTable } from 'ml-hash-table';
 70 | export { default as padArray } from 'ml-pad-array';
 71 | export { default as binarySearch } from 'binary-search';
 72 | export { Random } from 'ml-random';
 73 | 
 74 | import min from 'ml-array-min';
 75 | import max from 'ml-array-max';
 76 | import median from 'ml-array-median';
 77 | import mean from 'ml-array-mean';
 78 | import mode from 'ml-array-mode';
 79 | import normed from 'ml-array-normed';
 80 | import rescale from 'ml-array-rescale';
 81 | import sequentialFill from 'ml-array-sequential-fill';
 82 | import sum from 'ml-array-sum';
 83 | import standardDeviation from 'ml-array-standard-deviation';
 84 | import variance from 'ml-array-variance';
 85 | export const Array = {
 86 |   min,
 87 |   max,
 88 |   median,
 89 |   mean,
 90 |   mode,
 91 |   normed,
 92 |   rescale,
 93 |   sequentialFill,
 94 |   standardDeviation,
 95 |   sum,
 96 |   variance,
 97 | };
 98 | 
 99 | import centroidsMerge from 'ml-array-xy-centroids-merge';
100 | import closestX from 'ml-arrayxy-closestx';
101 | import covariance from 'ml-array-xy-covariance';
102 | import maxMerge from 'ml-array-xy-max-merge';
103 | import maxY from 'ml-array-xy-max-y';
104 | import sortX from 'ml-array-xy-sort-x';
105 | import uniqueX from 'ml-arrayxy-uniquex';
106 | import weightedMerge from 'ml-array-xy-weighted-merge';
107 | import equallySpaced from 'ml-array-xy-equally-spaced';
108 | import filterX from 'ml-array-xy-filter-x';
109 | export const ArrayXY = {
110 |   centroidsMerge,
111 |   closestX,
112 |   covariance,
113 |   maxMerge,
114 |   maxY,
115 |   sortX,
116 |   uniqueX,
117 |   weightedMerge,
118 |   equallySpaced,
119 |   filterX,
120 | };
121 | export { DecisionTreeClassifier, DecisionTreeRegression } from 'ml-cart';
122 | export {
123 |   RandomForestClassifier,
124 |   RandomForestRegression,
125 | } from 'ml-random-forest';
126 | export * as HClust from 'ml-hclust';
127 | export * as KMeans from 'ml-kmeans';
128 | export * as NaiveBayes from 'ml-naivebayes';
129 | export * as CrossValidation from 'ml-cross-validation';
130 | export * as FCNNLS from 'ml-fcnnls';
131 | export * as MatrixLib from 'ml-matrix';
132 | export * as GSD from 'ml-gsd';
133 | 


--------------------------------------------------------------------------------