@datagrok/eda 1.1.28 → 1.1.29

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (31) hide show
  1. package/CHANGELOG.md +5 -0
  2. package/README.md +1 -0
  3. package/dist/{05e5e0770f54f07e9474.wasm → 12a82b8001995d426ed2.wasm} +0 -0
  4. package/dist/23.js +1 -1
  5. package/dist/23.js.map +1 -1
  6. package/dist/501.js +2 -0
  7. package/dist/501.js.map +1 -0
  8. package/dist/727.js +2 -0
  9. package/dist/727.js.map +1 -0
  10. package/dist/package.js +1 -1
  11. package/dist/package.js.map +1 -1
  12. package/package.json +5 -5
  13. package/scripts/command.txt +1 -1
  14. package/scripts/func.json +664 -1
  15. package/scripts/module.json +1 -1
  16. package/src/data-generators.ts +1 -44
  17. package/src/missing-values-imputation/ui.ts +16 -6
  18. package/src/package.ts +60 -78
  19. package/src/regression.ts +1 -1
  20. package/src/softmax-classifier.ts +412 -0
  21. package/src/svm.ts +11 -33
  22. package/src/workers/softmax-worker.ts +146 -0
  23. package/wasm/EDA.js +55 -1
  24. package/wasm/EDA.wasm +0 -0
  25. package/wasm/EDAAPI.js +15 -0
  26. package/wasm/EDAForWebWorker.js +1 -1
  27. package/wasm/regression.h +2 -5
  28. package/wasm/softmax-api.cpp +49 -0
  29. package/wasm/softmax.h +156 -0
  30. package/wasm/workers/fitSoftmaxWorker.js +13 -0
  31. package/webpack.config.js +3 -2
package/scripts/func.json CHANGED
@@ -1 +1,664 @@
1
- {"pcaExport.cpp": {"principalComponentAnalysis": {"arguments": {"columns": {"type": "floatColumns"}, "componentsCount": {"type": "num"}, "centerNum": {"type": "num"}, "scaleNum": {"type": "num"}, "components": {"type": "newFloatColumns", "numOfRows": {"ref": "columns", "value": "numOfRows"}, "numOfColumns": {"ref": "componentsCount", "value": "data"}}}, "output": {"type": "tableFromColumns", "source": "components"}, "annotation": ["//name: principalComponentAnalysis", "//input: dataframe table", "//input: column_list columns", "//input: int componentsCount", "//input: int centerNum", "//input: int scaleNum", "//output: dataframe result "], "prototype": "principalComponentAnalysis(table, columns, componentsCount, centerNum, scaleNum)", "prototypeForWebWorker": "principalComponentAnalysisInWebWorker(table, columns, componentsCount, centerNum, scaleNum)", "callArgs": "[columns, componentsCount, centerNum, scaleNum]"}, "error": {"arguments": {"col1": {"type": "floatColumn"}, "col2": {"type": "floatColumn"}}, "output": {"type": "double", "source": "_callResult"}, "annotation": ["//name: error", "//input: dataframe df", "//input: column col1", "//input: column col2", "//output: double mad "], "prototype": "error(df, col1, col2)", "prototypeForWebWorker": "errorInWebWorker(df, col1, col2)", "callArgs": "[col1, col2]"}}, "PCA/PCA.cpp": {}, "plsExport.cpp": {"partialLeastSquareRegression": {"arguments": {"features": {"type": "floatColumns"}, "predict": {"type": "floatColumn"}, "componentsCount": {"type": "num"}, "prediction": {"type": "newFloatColumn", "numOfRows": {"ref": "predict", "value": "numOfRows"}}, "regressionCoefficients": {"type": "newFloatColumn", "numOfRows": {"ref": "features", "value": "numOfColumns"}}, "tScores": {"type": "newFloatColumns", "numOfRows": {"ref": "predict", "value": "numOfRows"}, "numOfColumns": {"ref": "componentsCount", "value": "data"}}, "uScores": {"type": "newFloatColumns", "numOfRows": {"ref": "predict", "value": "numOfRows"}, "numOfColumns": {"ref": "componentsCount", "value": "data"}}, "xLoadings": {"type": "newFloatColumns", "numOfRows": {"ref": "features", "value": "numOfColumns"}, "numOfColumns": {"ref": "componentsCount", "value": "data"}}, "yLoadings": {"type": "newFloatColumn", "numOfRows": {"ref": "componentsCount", "value": "data"}}}, "output": {"type": "objects", "source": "['prediction', 'regressionCoefficients', 'tScores', 'uScores', 'xLoadings', 'yLoadings']"}, "annotation": ["//name: partialLeastSquareRegression", "//input: dataframe table", "//input: column_list features", "//input: column predict", "//input: int componentsCount"], "prototype": "partialLeastSquareRegression(table, features, predict, componentsCount)", "prototypeForWebWorker": "partialLeastSquareRegressionInWebWorker(table, features, predict, componentsCount)", "callArgs": "[features, predict, componentsCount]"}}, "PLS/PLS.cpp": {}, "svmApi.cpp": {"generateDataset": {"arguments": {"kernel": {"type": "num"}, "kernelParams": {"type": "floatColumn"}, "samplesCount": {"type": "num"}, "featuresCount": {"type": "num"}, "min": {"type": "num"}, "max": {"type": "num"}, "violatorsPercentage": {"type": "num"}, "dataset": {"type": "newFloatColumns", "numOfRows": {"ref": "samplesCount", "value": "data"}, "numOfColumns": {"ref": "featuresCount", "value": "data"}}, "labels": {"type": "newFloatColumn", "numOfRows": {"ref": "samplesCount", "value": "data"}}}, "output": {"type": "objects", "source": "['dataset', 'labels']"}, "annotation": ["//name: generateDataset", "//input: int kernel", "//input: column kernelParams", "//input: int samplesCount", "//input: int featuresCount", "//input: double min", "//input: double max", "//input: double violatorsPercentage"], "prototype": "generateDataset(kernel, kernelParams, samplesCount, featuresCount, min, max, violatorsPercentage)", "prototypeForWebWorker": "generateDatasetInWebWorker(kernel, kernelParams, samplesCount, featuresCount, min, max, violatorsPercentage)", "callArgs": "[kernel, kernelParams, samplesCount, featuresCount, min, max, violatorsPercentage]"}, "normalizeDataset": {"arguments": {"data": {"type": "floatColumns"}, "normalizedData": {"type": "newFloatColumns", "numOfRows": {"ref": "data", "value": "numOfColumns"}, "numOfColumns": {"ref": "data", "value": "numOfRows"}}, "means": {"type": "newFloatColumn", "numOfRows": {"ref": "data", "value": "numOfColumns"}}, "stdDevs": {"type": "newFloatColumn", "numOfRows": {"ref": "data", "value": "numOfColumns"}}}, "output": {"type": "objects", "source": "['normalizedData', 'means', 'stdDevs']"}, "annotation": ["//name: normalizeDataset", "//input: column_list data"], "prototype": "normalizeDataset(data)", "prototypeForWebWorker": "normalizeDatasetInWebWorker(data)", "callArgs": "[data]"}, "trainLSSVM": {"arguments": {"gamma": {"type": "num"}, "kernel": {"type": "num"}, "kernelParams": {"type": "floatColumn"}, "modelParamsCount": {"type": "num"}, "precomputedWeightsCount": {"type": "num"}, "dataset": {"type": "floatColumns"}, "labels": {"type": "floatColumn"}, "normalizedData": {"type": "newFloatColumns", "numOfRows": {"ref": "dataset", "value": "numOfColumns"}, "numOfColumns": {"ref": "dataset", "value": "numOfRows"}}, "means": {"type": "newFloatColumn", "numOfRows": {"ref": "dataset", "value": "numOfColumns"}}, "stdDevs": {"type": "newFloatColumn", "numOfRows": {"ref": "dataset", "value": "numOfColumns"}}, "modelParams": {"type": "newFloatColumn", "numOfRows": {"ref": "modelParamsCount", "value": "data"}}, "precomputedWeights": {"type": "newFloatColumn", "numOfRows": {"ref": "precomputedWeightsCount", "value": "data"}}}, "output": {"type": "objects", "source": "['normalizedData', 'means', 'stdDevs', 'modelParams', 'precomputedWeights']"}, "annotation": ["//name: trainLSSVM", "//input: double gamma", "//input: int kernel", "//input: column kernelParams", "//input: int modelParamsCount", "//input: int precomputedWeightsCount", "//input: column_list dataset", "//input: column labels"], "prototype": "trainLSSVM(gamma, kernel, kernelParams, modelParamsCount, precomputedWeightsCount, dataset, labels)", "prototypeForWebWorker": "trainLSSVMInWebWorker(gamma, kernel, kernelParams, modelParamsCount, precomputedWeightsCount, dataset, labels)", "callArgs": "[gamma, kernel, kernelParams, modelParamsCount, precomputedWeightsCount, dataset, labels]"}, "predictByLSSVM": {"arguments": {"kernel": {"type": "num"}, "kernelParams": {"type": "floatColumn"}, "normalizedData": {"type": "floatColumns"}, "labels": {"type": "floatColumn"}, "means": {"type": "floatColumn"}, "stdDevs": {"type": "floatColumn"}, "modelParams": {"type": "floatColumn"}, "precomputedWeights": {"type": "floatColumn"}, "targetData": {"type": "floatColumns"}, "prediction": {"type": "newFloatColumn", "numOfRows": {"ref": "targetData", "value": "numOfRows"}}}, "output": {"type": "column", "source": "prediction"}, "annotation": ["//name: predictByLSSVM", "//input: int kernel", "//input: column kernelParams", "//input: column_list normalizedData", "//input: column labels", "//input: column means", "//input: column stdDevs", "//input: column modelParams", "//input: column precomputedWeights", "//input: column_list targetData", "//output: column prediction"], "prototype": "predictByLSSVM(kernel, kernelParams, normalizedData, labels, means, stdDevs, modelParams, precomputedWeights, targetData)", "prototypeForWebWorker": "predictByLSSVMInWebWorker(kernel, kernelParams, normalizedData, labels, means, stdDevs, modelParams, precomputedWeights, targetData)", "callArgs": "[kernel, kernelParams, normalizedData, labels, means, stdDevs, modelParams, precomputedWeights, targetData]"}, "trainAndAnalyzeLSSVM": {"arguments": {"gamma": {"type": "num"}, "kernel": {"type": "num"}, "kernelParams": {"type": "floatColumn"}, "modelParamsCount": {"type": "num"}, "precomputedWeightsCount": {"type": "num"}, "confusionMatrixElementsCount": {"type": "num"}, "dataset": {"type": "floatColumns"}, "labels": {"type": "floatColumn"}, "normalizedData": {"type": "newFloatColumns", "numOfRows": {"ref": "dataset", "value": "numOfColumns"}, "numOfColumns": {"ref": "dataset", "value": "numOfRows"}}, "means": {"type": "newFloatColumn", "numOfRows": {"ref": "dataset", "value": "numOfColumns"}}, "stdDevs": {"type": "newFloatColumn", "numOfRows": {"ref": "dataset", "value": "numOfColumns"}}, "modelParams": {"type": "newFloatColumn", "numOfRows": {"ref": "modelParamsCount", "value": "data"}}, "precomputedWeights": {"type": "newFloatColumn", "numOfRows": {"ref": "precomputedWeightsCount", "value": "data"}}, "predictedLabels": {"type": "newFloatColumn", "numOfRows": {"ref": "dataset", "value": "numOfRows"}}, "correctness": {"type": "newFloatColumn", "numOfRows": {"ref": "dataset", "value": "numOfRows"}}, "consfusionMatrix": {"type": "newIntColumn", "numOfRows": {"ref": "confusionMatrixElementsCount", "value": "data"}}}, "output": {"type": "objects", "source": "['normalizedData', 'means', 'stdDevs', 'modelParams', 'precomputedWeights', 'predictedLabels', 'correctness', 'consfusionMatrix']"}, "annotation": ["//name: trainAndAnalyzeLSSVM", "//input: double gamma", "//input: int kernel", "//input: column kernelParams", "//input: int modelParamsCount", "//input: int precomputedWeightsCount", "//input: int confusionMatrixElementsCount", "//input: column_list dataset", "//input: column labels"], "prototype": "trainAndAnalyzeLSSVM(gamma, kernel, kernelParams, modelParamsCount, precomputedWeightsCount, confusionMatrixElementsCount, dataset, labels)", "prototypeForWebWorker": "trainAndAnalyzeLSSVMInWebWorker(gamma, kernel, kernelParams, modelParamsCount, precomputedWeightsCount, confusionMatrixElementsCount, dataset, labels)", "callArgs": "[gamma, kernel, kernelParams, modelParamsCount, precomputedWeightsCount, confusionMatrixElementsCount, dataset, labels]"}}, "regression-api.cpp": {"fitLinearRegressionParamsWithDataNormalizing": {"arguments": {"features": {"type": "floatColumns"}, "featureAvgs": {"type": "floatColumn"}, "featureStdDevs": {"type": "floatColumn"}, "targets": {"type": "floatColumn"}, "targetsAvg": {"type": "num"}, "targetsStdDev": {"type": "num"}, "paramsCount": {"type": "num"}, "params": {"type": "newFloatColumn", "numOfRows": {"ref": "paramsCount", "value": "data"}}}, "output": {"type": "column", "source": "params"}, "annotation": ["//name: fitLinearRegressionParamsWithDataNormalizing", "//input: column_list features", "//input: column featureAvgs", "//input: column featureStdDevs", "//input: column targets", "//input: double targetsAvg", "//input: double targetsStdDev", "//input: int paramsCount", "//output: column params"], "prototype": "fitLinearRegressionParamsWithDataNormalizing(features, featureAvgs, featureStdDevs, targets, targetsAvg, targetsStdDev, paramsCount)", "prototypeForWebWorker": "fitLinearRegressionParamsWithDataNormalizingInWebWorker(features, featureAvgs, featureStdDevs, targets, targetsAvg, targetsStdDev, paramsCount)", "callArgs": "[features, featureAvgs, featureStdDevs, targets, targetsAvg, targetsStdDev, paramsCount]"}, "fitLinearRegressionParams": {"arguments": {"features": {"type": "floatColumns"}, "targets": {"type": "floatColumn"}, "paramsCount": {"type": "num"}, "params": {"type": "newFloatColumn", "numOfRows": {"ref": "paramsCount", "value": "data"}}}, "output": {"type": "column", "source": "params"}, "annotation": ["//name: fitLinearRegressionParams", "//input: column_list features", "//input: column targets", "//input: int paramsCount", "//output: column params"], "prototype": "fitLinearRegressionParams(features, targets, paramsCount)", "prototypeForWebWorker": "fitLinearRegressionParamsInWebWorker(features, targets, paramsCount)", "callArgs": "[features, targets, paramsCount]"}}}
1
+ {
2
+ "pcaExport.cpp": {
3
+ "principalComponentAnalysis": {
4
+ "arguments": {
5
+ "columns": {
6
+ "type": "floatColumns"
7
+ },
8
+ "componentsCount": {
9
+ "type": "num"
10
+ },
11
+ "centerNum": {
12
+ "type": "num"
13
+ },
14
+ "scaleNum": {
15
+ "type": "num"
16
+ },
17
+ "components": {
18
+ "type": "newFloatColumns",
19
+ "numOfRows": {
20
+ "ref": "columns",
21
+ "value": "numOfRows"
22
+ },
23
+ "numOfColumns": {
24
+ "ref": "componentsCount",
25
+ "value": "data"
26
+ }
27
+ }
28
+ },
29
+ "output": {
30
+ "type": "tableFromColumns",
31
+ "source": "components"
32
+ },
33
+ "annotation": [
34
+ "//name: principalComponentAnalysis",
35
+ "//input: dataframe table",
36
+ "//input: column_list columns",
37
+ "//input: int componentsCount",
38
+ "//input: int centerNum",
39
+ "//input: int scaleNum",
40
+ "//output: dataframe result "
41
+ ],
42
+ "prototype": "principalComponentAnalysis(table, columns, componentsCount, centerNum, scaleNum)",
43
+ "prototypeForWebWorker": "principalComponentAnalysisInWebWorker(table, columns, componentsCount, centerNum, scaleNum)",
44
+ "callArgs": "[columns, componentsCount, centerNum, scaleNum]"
45
+ },
46
+ "error": {
47
+ "arguments": {
48
+ "col1": {
49
+ "type": "floatColumn"
50
+ },
51
+ "col2": {
52
+ "type": "floatColumn"
53
+ }
54
+ },
55
+ "output": {
56
+ "type": "double",
57
+ "source": "_callResult"
58
+ },
59
+ "annotation": [
60
+ "//name: error",
61
+ "//input: dataframe df",
62
+ "//input: column col1",
63
+ "//input: column col2",
64
+ "//output: double mad "
65
+ ],
66
+ "prototype": "error(df, col1, col2)",
67
+ "prototypeForWebWorker": "errorInWebWorker(df, col1, col2)",
68
+ "callArgs": "[col1, col2]"
69
+ }
70
+ },
71
+ "PCA/PCA.cpp": {},
72
+ "plsExport.cpp": {
73
+ "partialLeastSquareRegression": {
74
+ "arguments": {
75
+ "features": {
76
+ "type": "floatColumns"
77
+ },
78
+ "predict": {
79
+ "type": "floatColumn"
80
+ },
81
+ "componentsCount": {
82
+ "type": "num"
83
+ },
84
+ "prediction": {
85
+ "type": "newFloatColumn",
86
+ "numOfRows": {
87
+ "ref": "predict",
88
+ "value": "numOfRows"
89
+ }
90
+ },
91
+ "regressionCoefficients": {
92
+ "type": "newFloatColumn",
93
+ "numOfRows": {
94
+ "ref": "features",
95
+ "value": "numOfColumns"
96
+ }
97
+ },
98
+ "tScores": {
99
+ "type": "newFloatColumns",
100
+ "numOfRows": {
101
+ "ref": "predict",
102
+ "value": "numOfRows"
103
+ },
104
+ "numOfColumns": {
105
+ "ref": "componentsCount",
106
+ "value": "data"
107
+ }
108
+ },
109
+ "uScores": {
110
+ "type": "newFloatColumns",
111
+ "numOfRows": {
112
+ "ref": "predict",
113
+ "value": "numOfRows"
114
+ },
115
+ "numOfColumns": {
116
+ "ref": "componentsCount",
117
+ "value": "data"
118
+ }
119
+ },
120
+ "xLoadings": {
121
+ "type": "newFloatColumns",
122
+ "numOfRows": {
123
+ "ref": "features",
124
+ "value": "numOfColumns"
125
+ },
126
+ "numOfColumns": {
127
+ "ref": "componentsCount",
128
+ "value": "data"
129
+ }
130
+ },
131
+ "yLoadings": {
132
+ "type": "newFloatColumn",
133
+ "numOfRows": {
134
+ "ref": "componentsCount",
135
+ "value": "data"
136
+ }
137
+ }
138
+ },
139
+ "output": {
140
+ "type": "objects",
141
+ "source": "['prediction', 'regressionCoefficients', 'tScores', 'uScores', 'xLoadings', 'yLoadings']"
142
+ },
143
+ "annotation": [
144
+ "//name: partialLeastSquareRegression",
145
+ "//input: dataframe table",
146
+ "//input: column_list features",
147
+ "//input: column predict",
148
+ "//input: int componentsCount"
149
+ ],
150
+ "prototype": "partialLeastSquareRegression(table, features, predict, componentsCount)",
151
+ "prototypeForWebWorker": "partialLeastSquareRegressionInWebWorker(table, features, predict, componentsCount)",
152
+ "callArgs": "[features, predict, componentsCount]"
153
+ }
154
+ },
155
+ "PLS/PLS.cpp": {},
156
+ "svmApi.cpp": {
157
+ "generateDataset": {
158
+ "arguments": {
159
+ "kernel": {
160
+ "type": "num"
161
+ },
162
+ "kernelParams": {
163
+ "type": "floatColumn"
164
+ },
165
+ "samplesCount": {
166
+ "type": "num"
167
+ },
168
+ "featuresCount": {
169
+ "type": "num"
170
+ },
171
+ "min": {
172
+ "type": "num"
173
+ },
174
+ "max": {
175
+ "type": "num"
176
+ },
177
+ "violatorsPercentage": {
178
+ "type": "num"
179
+ },
180
+ "dataset": {
181
+ "type": "newFloatColumns",
182
+ "numOfRows": {
183
+ "ref": "samplesCount",
184
+ "value": "data"
185
+ },
186
+ "numOfColumns": {
187
+ "ref": "featuresCount",
188
+ "value": "data"
189
+ }
190
+ },
191
+ "labels": {
192
+ "type": "newFloatColumn",
193
+ "numOfRows": {
194
+ "ref": "samplesCount",
195
+ "value": "data"
196
+ }
197
+ }
198
+ },
199
+ "output": {
200
+ "type": "objects",
201
+ "source": "['dataset', 'labels']"
202
+ },
203
+ "annotation": [
204
+ "//name: generateDataset",
205
+ "//input: int kernel",
206
+ "//input: column kernelParams",
207
+ "//input: int samplesCount",
208
+ "//input: int featuresCount",
209
+ "//input: double min",
210
+ "//input: double max",
211
+ "//input: double violatorsPercentage"
212
+ ],
213
+ "prototype": "generateDataset(kernel, kernelParams, samplesCount, featuresCount, min, max, violatorsPercentage)",
214
+ "prototypeForWebWorker": "generateDatasetInWebWorker(kernel, kernelParams, samplesCount, featuresCount, min, max, violatorsPercentage)",
215
+ "callArgs": "[kernel, kernelParams, samplesCount, featuresCount, min, max, violatorsPercentage]"
216
+ },
217
+ "normalizeDataset": {
218
+ "arguments": {
219
+ "data": {
220
+ "type": "floatColumns"
221
+ },
222
+ "normalizedData": {
223
+ "type": "newFloatColumns",
224
+ "numOfRows": {
225
+ "ref": "data",
226
+ "value": "numOfColumns"
227
+ },
228
+ "numOfColumns": {
229
+ "ref": "data",
230
+ "value": "numOfRows"
231
+ }
232
+ },
233
+ "means": {
234
+ "type": "newFloatColumn",
235
+ "numOfRows": {
236
+ "ref": "data",
237
+ "value": "numOfColumns"
238
+ }
239
+ },
240
+ "stdDevs": {
241
+ "type": "newFloatColumn",
242
+ "numOfRows": {
243
+ "ref": "data",
244
+ "value": "numOfColumns"
245
+ }
246
+ }
247
+ },
248
+ "output": {
249
+ "type": "objects",
250
+ "source": "['normalizedData', 'means', 'stdDevs']"
251
+ },
252
+ "annotation": [
253
+ "//name: normalizeDataset",
254
+ "//input: column_list data"
255
+ ],
256
+ "prototype": "normalizeDataset(data)",
257
+ "prototypeForWebWorker": "normalizeDatasetInWebWorker(data)",
258
+ "callArgs": "[data]"
259
+ },
260
+ "trainLSSVM": {
261
+ "arguments": {
262
+ "gamma": {
263
+ "type": "num"
264
+ },
265
+ "kernel": {
266
+ "type": "num"
267
+ },
268
+ "kernelParams": {
269
+ "type": "floatColumn"
270
+ },
271
+ "modelParamsCount": {
272
+ "type": "num"
273
+ },
274
+ "precomputedWeightsCount": {
275
+ "type": "num"
276
+ },
277
+ "dataset": {
278
+ "type": "floatColumns"
279
+ },
280
+ "labels": {
281
+ "type": "floatColumn"
282
+ },
283
+ "normalizedData": {
284
+ "type": "newFloatColumns",
285
+ "numOfRows": {
286
+ "ref": "dataset",
287
+ "value": "numOfColumns"
288
+ },
289
+ "numOfColumns": {
290
+ "ref": "dataset",
291
+ "value": "numOfRows"
292
+ }
293
+ },
294
+ "means": {
295
+ "type": "newFloatColumn",
296
+ "numOfRows": {
297
+ "ref": "dataset",
298
+ "value": "numOfColumns"
299
+ }
300
+ },
301
+ "stdDevs": {
302
+ "type": "newFloatColumn",
303
+ "numOfRows": {
304
+ "ref": "dataset",
305
+ "value": "numOfColumns"
306
+ }
307
+ },
308
+ "modelParams": {
309
+ "type": "newFloatColumn",
310
+ "numOfRows": {
311
+ "ref": "modelParamsCount",
312
+ "value": "data"
313
+ }
314
+ },
315
+ "precomputedWeights": {
316
+ "type": "newFloatColumn",
317
+ "numOfRows": {
318
+ "ref": "precomputedWeightsCount",
319
+ "value": "data"
320
+ }
321
+ }
322
+ },
323
+ "output": {
324
+ "type": "objects",
325
+ "source": "['normalizedData', 'means', 'stdDevs', 'modelParams', 'precomputedWeights']"
326
+ },
327
+ "annotation": [
328
+ "//name: trainLSSVM",
329
+ "//input: double gamma",
330
+ "//input: int kernel",
331
+ "//input: column kernelParams",
332
+ "//input: int modelParamsCount",
333
+ "//input: int precomputedWeightsCount",
334
+ "//input: column_list dataset",
335
+ "//input: column labels"
336
+ ],
337
+ "prototype": "trainLSSVM(gamma, kernel, kernelParams, modelParamsCount, precomputedWeightsCount, dataset, labels)",
338
+ "prototypeForWebWorker": "trainLSSVMInWebWorker(gamma, kernel, kernelParams, modelParamsCount, precomputedWeightsCount, dataset, labels)",
339
+ "callArgs": "[gamma, kernel, kernelParams, modelParamsCount, precomputedWeightsCount, dataset, labels]"
340
+ },
341
+ "predictByLSSVM": {
342
+ "arguments": {
343
+ "kernel": {
344
+ "type": "num"
345
+ },
346
+ "kernelParams": {
347
+ "type": "floatColumn"
348
+ },
349
+ "normalizedData": {
350
+ "type": "floatColumns"
351
+ },
352
+ "labels": {
353
+ "type": "floatColumn"
354
+ },
355
+ "means": {
356
+ "type": "floatColumn"
357
+ },
358
+ "stdDevs": {
359
+ "type": "floatColumn"
360
+ },
361
+ "modelParams": {
362
+ "type": "floatColumn"
363
+ },
364
+ "precomputedWeights": {
365
+ "type": "floatColumn"
366
+ },
367
+ "targetData": {
368
+ "type": "floatColumns"
369
+ },
370
+ "prediction": {
371
+ "type": "newFloatColumn",
372
+ "numOfRows": {
373
+ "ref": "targetData",
374
+ "value": "numOfRows"
375
+ }
376
+ }
377
+ },
378
+ "output": {
379
+ "type": "column",
380
+ "source": "prediction"
381
+ },
382
+ "annotation": [
383
+ "//name: predictByLSSVM",
384
+ "//input: int kernel",
385
+ "//input: column kernelParams",
386
+ "//input: column_list normalizedData",
387
+ "//input: column labels",
388
+ "//input: column means",
389
+ "//input: column stdDevs",
390
+ "//input: column modelParams",
391
+ "//input: column precomputedWeights",
392
+ "//input: column_list targetData",
393
+ "//output: column prediction"
394
+ ],
395
+ "prototype": "predictByLSSVM(kernel, kernelParams, normalizedData, labels, means, stdDevs, modelParams, precomputedWeights, targetData)",
396
+ "prototypeForWebWorker": "predictByLSSVMInWebWorker(kernel, kernelParams, normalizedData, labels, means, stdDevs, modelParams, precomputedWeights, targetData)",
397
+ "callArgs": "[kernel, kernelParams, normalizedData, labels, means, stdDevs, modelParams, precomputedWeights, targetData]"
398
+ },
399
+ "trainAndAnalyzeLSSVM": {
400
+ "arguments": {
401
+ "gamma": {
402
+ "type": "num"
403
+ },
404
+ "kernel": {
405
+ "type": "num"
406
+ },
407
+ "kernelParams": {
408
+ "type": "floatColumn"
409
+ },
410
+ "modelParamsCount": {
411
+ "type": "num"
412
+ },
413
+ "precomputedWeightsCount": {
414
+ "type": "num"
415
+ },
416
+ "confusionMatrixElementsCount": {
417
+ "type": "num"
418
+ },
419
+ "dataset": {
420
+ "type": "floatColumns"
421
+ },
422
+ "labels": {
423
+ "type": "floatColumn"
424
+ },
425
+ "normalizedData": {
426
+ "type": "newFloatColumns",
427
+ "numOfRows": {
428
+ "ref": "dataset",
429
+ "value": "numOfColumns"
430
+ },
431
+ "numOfColumns": {
432
+ "ref": "dataset",
433
+ "value": "numOfRows"
434
+ }
435
+ },
436
+ "means": {
437
+ "type": "newFloatColumn",
438
+ "numOfRows": {
439
+ "ref": "dataset",
440
+ "value": "numOfColumns"
441
+ }
442
+ },
443
+ "stdDevs": {
444
+ "type": "newFloatColumn",
445
+ "numOfRows": {
446
+ "ref": "dataset",
447
+ "value": "numOfColumns"
448
+ }
449
+ },
450
+ "modelParams": {
451
+ "type": "newFloatColumn",
452
+ "numOfRows": {
453
+ "ref": "modelParamsCount",
454
+ "value": "data"
455
+ }
456
+ },
457
+ "precomputedWeights": {
458
+ "type": "newFloatColumn",
459
+ "numOfRows": {
460
+ "ref": "precomputedWeightsCount",
461
+ "value": "data"
462
+ }
463
+ },
464
+ "predictedLabels": {
465
+ "type": "newFloatColumn",
466
+ "numOfRows": {
467
+ "ref": "dataset",
468
+ "value": "numOfRows"
469
+ }
470
+ },
471
+ "correctness": {
472
+ "type": "newFloatColumn",
473
+ "numOfRows": {
474
+ "ref": "dataset",
475
+ "value": "numOfRows"
476
+ }
477
+ },
478
+ "consfusionMatrix": {
479
+ "type": "newIntColumn",
480
+ "numOfRows": {
481
+ "ref": "confusionMatrixElementsCount",
482
+ "value": "data"
483
+ }
484
+ }
485
+ },
486
+ "output": {
487
+ "type": "objects",
488
+ "source": "['normalizedData', 'means', 'stdDevs', 'modelParams', 'precomputedWeights', 'predictedLabels', 'correctness', 'consfusionMatrix']"
489
+ },
490
+ "annotation": [
491
+ "//name: trainAndAnalyzeLSSVM",
492
+ "//input: double gamma",
493
+ "//input: int kernel",
494
+ "//input: column kernelParams",
495
+ "//input: int modelParamsCount",
496
+ "//input: int precomputedWeightsCount",
497
+ "//input: int confusionMatrixElementsCount",
498
+ "//input: column_list dataset",
499
+ "//input: column labels"
500
+ ],
501
+ "prototype": "trainAndAnalyzeLSSVM(gamma, kernel, kernelParams, modelParamsCount, precomputedWeightsCount, confusionMatrixElementsCount, dataset, labels)",
502
+ "prototypeForWebWorker": "trainAndAnalyzeLSSVMInWebWorker(gamma, kernel, kernelParams, modelParamsCount, precomputedWeightsCount, confusionMatrixElementsCount, dataset, labels)",
503
+ "callArgs": "[gamma, kernel, kernelParams, modelParamsCount, precomputedWeightsCount, confusionMatrixElementsCount, dataset, labels]"
504
+ }
505
+ },
506
+ "regression-api.cpp": {
507
+ "fitLinearRegressionParamsWithDataNormalizing": {
508
+ "arguments": {
509
+ "features": {
510
+ "type": "floatColumns"
511
+ },
512
+ "featureAvgs": {
513
+ "type": "floatColumn"
514
+ },
515
+ "featureStdDevs": {
516
+ "type": "floatColumn"
517
+ },
518
+ "targets": {
519
+ "type": "floatColumn"
520
+ },
521
+ "targetsAvg": {
522
+ "type": "num"
523
+ },
524
+ "targetsStdDev": {
525
+ "type": "num"
526
+ },
527
+ "paramsCount": {
528
+ "type": "num"
529
+ },
530
+ "params": {
531
+ "type": "newFloatColumn",
532
+ "numOfRows": {
533
+ "ref": "paramsCount",
534
+ "value": "data"
535
+ }
536
+ }
537
+ },
538
+ "output": {
539
+ "type": "column",
540
+ "source": "params"
541
+ },
542
+ "annotation": [
543
+ "//name: fitLinearRegressionParamsWithDataNormalizing",
544
+ "//input: column_list features",
545
+ "//input: column featureAvgs",
546
+ "//input: column featureStdDevs",
547
+ "//input: column targets",
548
+ "//input: double targetsAvg",
549
+ "//input: double targetsStdDev",
550
+ "//input: int paramsCount",
551
+ "//output: column params"
552
+ ],
553
+ "prototype": "fitLinearRegressionParamsWithDataNormalizing(features, featureAvgs, featureStdDevs, targets, targetsAvg, targetsStdDev, paramsCount)",
554
+ "prototypeForWebWorker": "fitLinearRegressionParamsWithDataNormalizingInWebWorker(features, featureAvgs, featureStdDevs, targets, targetsAvg, targetsStdDev, paramsCount)",
555
+ "callArgs": "[features, featureAvgs, featureStdDevs, targets, targetsAvg, targetsStdDev, paramsCount]"
556
+ },
557
+ "fitLinearRegressionParams": {
558
+ "arguments": {
559
+ "features": {
560
+ "type": "floatColumns"
561
+ },
562
+ "targets": {
563
+ "type": "floatColumn"
564
+ },
565
+ "paramsCount": {
566
+ "type": "num"
567
+ },
568
+ "params": {
569
+ "type": "newFloatColumn",
570
+ "numOfRows": {
571
+ "ref": "paramsCount",
572
+ "value": "data"
573
+ }
574
+ }
575
+ },
576
+ "output": {
577
+ "type": "column",
578
+ "source": "params"
579
+ },
580
+ "annotation": [
581
+ "//name: fitLinearRegressionParams",
582
+ "//input: column_list features",
583
+ "//input: column targets",
584
+ "//input: int paramsCount",
585
+ "//output: column params"
586
+ ],
587
+ "prototype": "fitLinearRegressionParams(features, targets, paramsCount)",
588
+ "prototypeForWebWorker": "fitLinearRegressionParamsInWebWorker(features, targets, paramsCount)",
589
+ "callArgs": "[features, targets, paramsCount]"
590
+ }
591
+ },
592
+ "softmax-api.cpp": {
593
+ "fitSoftmax": {
594
+ "arguments": {
595
+ "features": {
596
+ "type": "floatColumns"
597
+ },
598
+ "featureAvgs": {
599
+ "type": "floatColumn"
600
+ },
601
+ "featureStdDevs": {
602
+ "type": "floatColumn"
603
+ },
604
+ "targets": {
605
+ "type": "intColumn"
606
+ },
607
+ "classesCount": {
608
+ "type": "num"
609
+ },
610
+ "iterCount": {
611
+ "type": "num"
612
+ },
613
+ "learningRate": {
614
+ "type": "num"
615
+ },
616
+ "penalty": {
617
+ "type": "num"
618
+ },
619
+ "tolerance": {
620
+ "type": "num"
621
+ },
622
+ "paramsRows": {
623
+ "type": "num"
624
+ },
625
+ "paramsCols": {
626
+ "type": "num"
627
+ },
628
+ "params": {
629
+ "type": "newFloatColumns",
630
+ "numOfRows": {
631
+ "ref": "paramsRows",
632
+ "value": "data"
633
+ },
634
+ "numOfColumns": {
635
+ "ref": "paramsCols",
636
+ "value": "data"
637
+ }
638
+ }
639
+ },
640
+ "output": {
641
+ "type": "tableFromColumns",
642
+ "source": "params"
643
+ },
644
+ "annotation": [
645
+ "//name: fitSoftmax",
646
+ "//input: column_list features",
647
+ "//input: column featureAvgs",
648
+ "//input: column featureStdDevs",
649
+ "//input: column targets",
650
+ "//input: int classesCount",
651
+ "//input: int iterCount",
652
+ "//input: double learningRate",
653
+ "//input: double penalty",
654
+ "//input: double tolerance",
655
+ "//input: int paramsRows",
656
+ "//input: int paramsCols",
657
+ "//output: dataframe result "
658
+ ],
659
+ "prototype": "fitSoftmax(features, featureAvgs, featureStdDevs, targets, classesCount, iterCount, learningRate, penalty, tolerance, paramsRows, paramsCols)",
660
+ "prototypeForWebWorker": "fitSoftmaxInWebWorker(features, featureAvgs, featureStdDevs, targets, classesCount, iterCount, learningRate, penalty, tolerance, paramsRows, paramsCols)",
661
+ "callArgs": "[features, featureAvgs, featureStdDevs, targets, classesCount, iterCount, learningRate, penalty, tolerance, paramsRows, paramsCols]"
662
+ }
663
+ }
664
+ }