1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
|
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<!-- NewPage -->
<html lang="en">
<head>
<!-- Generated by javadoc (version 1.7.0_51) on Wed Sep 16 15:55:12 PDT 2015 -->
<title>GradientDescent</title>
<meta name="date" content="2015-09-16">
<link rel="stylesheet" type="text/css" href="../../../../../stylesheet.css" title="Style">
</head>
<body>
<script type="text/javascript"><!--
if (location.href.indexOf('is-external=true') == -1) {
parent.document.title="GradientDescent";
}
//-->
</script>
<noscript>
<div>JavaScript is disabled on your browser.</div>
</noscript>
<!-- ========= START OF TOP NAVBAR ======= -->
<div class="topNav"><a name="navbar_top">
<!-- -->
</a><a href="#skip-navbar_top" title="Skip navigation links"></a><a name="navbar_top_firstrow">
<!-- -->
</a>
<ul class="navList" title="Navigation">
<li><a href="../../../../../overview-summary.html">Overview</a></li>
<li><a href="package-summary.html">Package</a></li>
<li class="navBarCell1Rev">Class</li>
<li><a href="package-tree.html">Tree</a></li>
<li><a href="../../../../../deprecated-list.html">Deprecated</a></li>
<li><a href="../../../../../index-all.html">Index</a></li>
<li><a href="../../../../../help-doc.html">Help</a></li>
</ul>
</div>
<div class="subNav">
<ul class="navList">
<li><a href="../../../../../org/apache/spark/mllib/optimization/Gradient.html" title="class in org.apache.spark.mllib.optimization"><span class="strong">Prev Class</span></a></li>
<li><a href="../../../../../org/apache/spark/mllib/optimization/HingeGradient.html" title="class in org.apache.spark.mllib.optimization"><span class="strong">Next Class</span></a></li>
</ul>
<ul class="navList">
<li><a href="../../../../../index.html?org/apache/spark/mllib/optimization/GradientDescent.html" target="_top">Frames</a></li>
<li><a href="GradientDescent.html" target="_top">No Frames</a></li>
</ul>
<ul class="navList" id="allclasses_navbar_top">
<li><a href="../../../../../allclasses-noframe.html">All Classes</a></li>
</ul>
<div>
<script type="text/javascript"><!--
allClassesLink = document.getElementById("allclasses_navbar_top");
if(window==top) {
allClassesLink.style.display = "block";
}
else {
allClassesLink.style.display = "none";
}
//-->
</script>
</div>
<div>
<ul class="subNavList">
<li>Summary: </li>
<li>Nested | </li>
<li>Field | </li>
<li>Constr | </li>
<li><a href="#method_summary">Method</a></li>
</ul>
<ul class="subNavList">
<li>Detail: </li>
<li>Field | </li>
<li>Constr | </li>
<li><a href="#method_detail">Method</a></li>
</ul>
</div>
<a name="skip-navbar_top">
<!-- -->
</a></div>
<!-- ========= END OF TOP NAVBAR ========= -->
<!-- ======== START OF CLASS DATA ======== -->
<div class="header">
<div class="subTitle">org.apache.spark.mllib.optimization</div>
<h2 title="Class GradientDescent" class="title">Class GradientDescent</h2>
</div>
<div class="contentContainer">
<ul class="inheritance">
<li>java.lang.Object</li>
<li>
<ul class="inheritance">
<li>org.apache.spark.mllib.optimization.GradientDescent</li>
</ul>
</li>
</ul>
<div class="description">
<ul class="blockList">
<li class="blockList">
<dl>
<dt>All Implemented Interfaces:</dt>
<dd>java.io.Serializable, <a href="../../../../../org/apache/spark/Logging.html" title="interface in org.apache.spark">Logging</a>, <a href="../../../../../org/apache/spark/mllib/optimization/Optimizer.html" title="interface in org.apache.spark.mllib.optimization">Optimizer</a></dd>
</dl>
<hr>
<br>
<pre>public class <span class="strong">GradientDescent</span>
extends java.lang.Object
implements <a href="../../../../../org/apache/spark/mllib/optimization/Optimizer.html" title="interface in org.apache.spark.mllib.optimization">Optimizer</a>, <a href="../../../../../org/apache/spark/Logging.html" title="interface in org.apache.spark">Logging</a></pre>
<div class="block">Class used to solve an optimization problem using Gradient Descent.
param: gradient Gradient function to be used.
param: updater Updater to be used to update weights after every iteration.</div>
<dl><dt><span class="strong">See Also:</span></dt><dd><a href="../../../../../serialized-form.html#org.apache.spark.mllib.optimization.GradientDescent">Serialized Form</a></dd></dl>
</li>
</ul>
</div>
<div class="summary">
<ul class="blockList">
<li class="blockList">
<!-- ========== METHOD SUMMARY =========== -->
<ul class="blockList">
<li class="blockList"><a name="method_summary">
<!-- -->
</a>
<h3>Method Summary</h3>
<table class="overviewSummary" border="0" cellpadding="3" cellspacing="0" summary="Method Summary table, listing methods, and an explanation">
<caption><span>Methods</span><span class="tabEnd"> </span></caption>
<tr>
<th class="colFirst" scope="col">Modifier and Type</th>
<th class="colLast" scope="col">Method and Description</th>
</tr>
<tr class="altColor">
<td class="colFirst"><code><a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a></code></td>
<td class="colLast"><code><strong><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html#optimize(org.apache.spark.rdd.RDD, org.apache.spark.mllib.linalg.Vector)">optimize</a></strong>(<a href="../../../../../org/apache/spark/rdd/RDD.html" title="class in org.apache.spark.rdd">RDD</a><scala.Tuple2<java.lang.Object,<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a>>> data,
<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a> initialWeights)</code>
<div class="block">:: DeveloperApi ::
Runs gradient descent on the given training data.</div>
</td>
</tr>
<tr class="rowColor">
<td class="colFirst"><code>static scala.Tuple2<<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a>,double[]></code></td>
<td class="colLast"><code><strong><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html#runMiniBatchSGD(org.apache.spark.rdd.RDD, org.apache.spark.mllib.optimization.Gradient, org.apache.spark.mllib.optimization.Updater, double, int, double, double, org.apache.spark.mllib.linalg.Vector)">runMiniBatchSGD</a></strong>(<a href="../../../../../org/apache/spark/rdd/RDD.html" title="class in org.apache.spark.rdd">RDD</a><scala.Tuple2<java.lang.Object,<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a>>> data,
<a href="../../../../../org/apache/spark/mllib/optimization/Gradient.html" title="class in org.apache.spark.mllib.optimization">Gradient</a> gradient,
<a href="../../../../../org/apache/spark/mllib/optimization/Updater.html" title="class in org.apache.spark.mllib.optimization">Updater</a> updater,
double stepSize,
int numIterations,
double regParam,
double miniBatchFraction,
<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a> initialWeights)</code>
<div class="block">Alias of <code>runMiniBatchSGD</code> with convergenceTol set to default value of 0.001.</div>
</td>
</tr>
<tr class="altColor">
<td class="colFirst"><code>static scala.Tuple2<<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a>,double[]></code></td>
<td class="colLast"><code><strong><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html#runMiniBatchSGD(org.apache.spark.rdd.RDD, org.apache.spark.mllib.optimization.Gradient, org.apache.spark.mllib.optimization.Updater, double, int, double, double, org.apache.spark.mllib.linalg.Vector, double)">runMiniBatchSGD</a></strong>(<a href="../../../../../org/apache/spark/rdd/RDD.html" title="class in org.apache.spark.rdd">RDD</a><scala.Tuple2<java.lang.Object,<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a>>> data,
<a href="../../../../../org/apache/spark/mllib/optimization/Gradient.html" title="class in org.apache.spark.mllib.optimization">Gradient</a> gradient,
<a href="../../../../../org/apache/spark/mllib/optimization/Updater.html" title="class in org.apache.spark.mllib.optimization">Updater</a> updater,
double stepSize,
int numIterations,
double regParam,
double miniBatchFraction,
<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a> initialWeights,
double convergenceTol)</code>
<div class="block">Run stochastic gradient descent (SGD) in parallel using mini batches.</div>
</td>
</tr>
<tr class="rowColor">
<td class="colFirst"><code><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a></code></td>
<td class="colLast"><code><strong><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html#setConvergenceTol(double)">setConvergenceTol</a></strong>(double tolerance)</code>
<div class="block">Set the convergence tolerance.</div>
</td>
</tr>
<tr class="altColor">
<td class="colFirst"><code><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a></code></td>
<td class="colLast"><code><strong><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html#setGradient(org.apache.spark.mllib.optimization.Gradient)">setGradient</a></strong>(<a href="../../../../../org/apache/spark/mllib/optimization/Gradient.html" title="class in org.apache.spark.mllib.optimization">Gradient</a> gradient)</code>
<div class="block">Set the gradient function (of the loss function of one single data example)
to be used for SGD.</div>
</td>
</tr>
<tr class="rowColor">
<td class="colFirst"><code><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a></code></td>
<td class="colLast"><code><strong><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html#setMiniBatchFraction(double)">setMiniBatchFraction</a></strong>(double fraction)</code>
<div class="block">:: Experimental ::
Set fraction of data to be used for each SGD iteration.</div>
</td>
</tr>
<tr class="altColor">
<td class="colFirst"><code><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a></code></td>
<td class="colLast"><code><strong><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html#setNumIterations(int)">setNumIterations</a></strong>(int iters)</code>
<div class="block">Set the number of iterations for SGD.</div>
</td>
</tr>
<tr class="rowColor">
<td class="colFirst"><code><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a></code></td>
<td class="colLast"><code><strong><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html#setRegParam(double)">setRegParam</a></strong>(double regParam)</code>
<div class="block">Set the regularization parameter.</div>
</td>
</tr>
<tr class="altColor">
<td class="colFirst"><code><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a></code></td>
<td class="colLast"><code><strong><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html#setStepSize(double)">setStepSize</a></strong>(double step)</code>
<div class="block">Set the initial step size of SGD for the first step.</div>
</td>
</tr>
<tr class="rowColor">
<td class="colFirst"><code><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a></code></td>
<td class="colLast"><code><strong><a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html#setUpdater(org.apache.spark.mllib.optimization.Updater)">setUpdater</a></strong>(<a href="../../../../../org/apache/spark/mllib/optimization/Updater.html" title="class in org.apache.spark.mllib.optimization">Updater</a> updater)</code>
<div class="block">Set the updater function to actually perform a gradient step in a given direction.</div>
</td>
</tr>
</table>
<ul class="blockList">
<li class="blockList"><a name="methods_inherited_from_class_java.lang.Object">
<!-- -->
</a>
<h3>Methods inherited from class java.lang.Object</h3>
<code>clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait</code></li>
</ul>
<ul class="blockList">
<li class="blockList"><a name="methods_inherited_from_class_org.apache.spark.Logging">
<!-- -->
</a>
<h3>Methods inherited from interface org.apache.spark.<a href="../../../../../org/apache/spark/Logging.html" title="interface in org.apache.spark">Logging</a></h3>
<code><a href="../../../../../org/apache/spark/Logging.html#initializeIfNecessary()">initializeIfNecessary</a>, <a href="../../../../../org/apache/spark/Logging.html#initializeLogging()">initializeLogging</a>, <a href="../../../../../org/apache/spark/Logging.html#isTraceEnabled()">isTraceEnabled</a>, <a href="../../../../../org/apache/spark/Logging.html#log_()">log_</a>, <a href="../../../../../org/apache/spark/Logging.html#log()">log</a>, <a href="../../../../../org/apache/spark/Logging.html#logDebug(scala.Function0)">logDebug</a>, <a href="../../../../../org/apache/spark/Logging.html#logDebug(scala.Function0, java.lang.Throwable)">logDebug</a>, <a href="../../../../../org/apache/spark/Logging.html#logError(scala.Function0)">logError</a>, <a href="../../../../../org/apache/spark/Logging.html#logError(scala.Function0, java.lang.Throwable)">logError</a>, <a href="../../../../../org/apache/spark/Logging.html#logInfo(scala.Function0)">logInfo</a>, <a href="../../../../../org/apache/spark/Logging.html#logInfo(scala.Function0, java.lang.Throwable)">logInfo</a>, <a href="../../../../../org/apache/spark/Logging.html#logName()">logName</a>, <a href="../../../../../org/apache/spark/Logging.html#logTrace(scala.Function0)">logTrace</a>, <a href="../../../../../org/apache/spark/Logging.html#logTrace(scala.Function0, java.lang.Throwable)">logTrace</a>, <a href="../../../../../org/apache/spark/Logging.html#logWarning(scala.Function0)">logWarning</a>, <a href="../../../../../org/apache/spark/Logging.html#logWarning(scala.Function0, java.lang.Throwable)">logWarning</a></code></li>
</ul>
</li>
</ul>
</li>
</ul>
</div>
<div class="details">
<ul class="blockList">
<li class="blockList">
<!-- ============ METHOD DETAIL ========== -->
<ul class="blockList">
<li class="blockList"><a name="method_detail">
<!-- -->
</a>
<h3>Method Detail</h3>
<a name="runMiniBatchSGD(org.apache.spark.rdd.RDD, org.apache.spark.mllib.optimization.Gradient, org.apache.spark.mllib.optimization.Updater, double, int, double, double, org.apache.spark.mllib.linalg.Vector, double)">
<!-- -->
</a>
<ul class="blockList">
<li class="blockList">
<h4>runMiniBatchSGD</h4>
<pre>public static scala.Tuple2<<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a>,double[]> runMiniBatchSGD(<a href="../../../../../org/apache/spark/rdd/RDD.html" title="class in org.apache.spark.rdd">RDD</a><scala.Tuple2<java.lang.Object,<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a>>> data,
<a href="../../../../../org/apache/spark/mllib/optimization/Gradient.html" title="class in org.apache.spark.mllib.optimization">Gradient</a> gradient,
<a href="../../../../../org/apache/spark/mllib/optimization/Updater.html" title="class in org.apache.spark.mllib.optimization">Updater</a> updater,
double stepSize,
int numIterations,
double regParam,
double miniBatchFraction,
<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a> initialWeights,
double convergenceTol)</pre>
<div class="block">Run stochastic gradient descent (SGD) in parallel using mini batches.
In each iteration, we sample a subset (fraction miniBatchFraction) of the total data
in order to compute a gradient estimate.
Sampling, and averaging the subgradients over this subset is performed using one standard
spark map-reduce in each iteration.
<p></div>
<dl><dt><span class="strong">Parameters:</span></dt><dd><code>data</code> - Input data for SGD. RDD of the set of data examples, each of
the form (label, [feature values]).</dd><dd><code>gradient</code> - Gradient object (used to compute the gradient of the loss function of
one single data example)</dd><dd><code>updater</code> - Updater function to actually perform a gradient step in a given direction.</dd><dd><code>stepSize</code> - initial step size for the first step</dd><dd><code>numIterations</code> - number of iterations that SGD should be run.</dd><dd><code>regParam</code> - regularization parameter</dd><dd><code>miniBatchFraction</code> - fraction of the input data set that should be used for
one iteration of SGD. Default value 1.0.</dd><dd><code>convergenceTol</code> - Minibatch iteration will end before numIterations if the relative
difference between the current weight and the previous weight is less
than this value. In measuring convergence, L2 norm is calculated.
Default value 0.001. Must be between 0.0 and 1.0 inclusively.</dd><dd><code>initialWeights</code> - (undocumented)</dd>
<dt><span class="strong">Returns:</span></dt><dd>A tuple containing two elements. The first element is a column matrix containing
weights for every feature, and the second element is an array containing the
stochastic loss computed for every iteration.</dd></dl>
</li>
</ul>
<a name="runMiniBatchSGD(org.apache.spark.rdd.RDD, org.apache.spark.mllib.optimization.Gradient, org.apache.spark.mllib.optimization.Updater, double, int, double, double, org.apache.spark.mllib.linalg.Vector)">
<!-- -->
</a>
<ul class="blockList">
<li class="blockList">
<h4>runMiniBatchSGD</h4>
<pre>public static scala.Tuple2<<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a>,double[]> runMiniBatchSGD(<a href="../../../../../org/apache/spark/rdd/RDD.html" title="class in org.apache.spark.rdd">RDD</a><scala.Tuple2<java.lang.Object,<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a>>> data,
<a href="../../../../../org/apache/spark/mllib/optimization/Gradient.html" title="class in org.apache.spark.mllib.optimization">Gradient</a> gradient,
<a href="../../../../../org/apache/spark/mllib/optimization/Updater.html" title="class in org.apache.spark.mllib.optimization">Updater</a> updater,
double stepSize,
int numIterations,
double regParam,
double miniBatchFraction,
<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a> initialWeights)</pre>
<div class="block">Alias of <code>runMiniBatchSGD</code> with convergenceTol set to default value of 0.001.</div>
<dl><dt><span class="strong">Parameters:</span></dt><dd><code>data</code> - (undocumented)</dd><dd><code>gradient</code> - (undocumented)</dd><dd><code>updater</code> - (undocumented)</dd><dd><code>stepSize</code> - (undocumented)</dd><dd><code>numIterations</code> - (undocumented)</dd><dd><code>regParam</code> - (undocumented)</dd><dd><code>miniBatchFraction</code> - (undocumented)</dd><dd><code>initialWeights</code> - (undocumented)</dd>
<dt><span class="strong">Returns:</span></dt><dd>(undocumented)</dd></dl>
</li>
</ul>
<a name="setStepSize(double)">
<!-- -->
</a>
<ul class="blockList">
<li class="blockList">
<h4>setStepSize</h4>
<pre>public <a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a> setStepSize(double step)</pre>
<div class="block">Set the initial step size of SGD for the first step. Default 1.0.
In subsequent steps, the step size will decrease with stepSize/sqrt(t)</div>
<dl><dt><span class="strong">Parameters:</span></dt><dd><code>step</code> - (undocumented)</dd>
<dt><span class="strong">Returns:</span></dt><dd>(undocumented)</dd></dl>
</li>
</ul>
<a name="setMiniBatchFraction(double)">
<!-- -->
</a>
<ul class="blockList">
<li class="blockList">
<h4>setMiniBatchFraction</h4>
<pre>public <a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a> setMiniBatchFraction(double fraction)</pre>
<div class="block">:: Experimental ::
Set fraction of data to be used for each SGD iteration.
Default 1.0 (corresponding to deterministic/classical gradient descent)</div>
<dl><dt><span class="strong">Parameters:</span></dt><dd><code>fraction</code> - (undocumented)</dd>
<dt><span class="strong">Returns:</span></dt><dd>(undocumented)</dd></dl>
</li>
</ul>
<a name="setNumIterations(int)">
<!-- -->
</a>
<ul class="blockList">
<li class="blockList">
<h4>setNumIterations</h4>
<pre>public <a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a> setNumIterations(int iters)</pre>
<div class="block">Set the number of iterations for SGD. Default 100.</div>
<dl><dt><span class="strong">Parameters:</span></dt><dd><code>iters</code> - (undocumented)</dd>
<dt><span class="strong">Returns:</span></dt><dd>(undocumented)</dd></dl>
</li>
</ul>
<a name="setRegParam(double)">
<!-- -->
</a>
<ul class="blockList">
<li class="blockList">
<h4>setRegParam</h4>
<pre>public <a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a> setRegParam(double regParam)</pre>
<div class="block">Set the regularization parameter. Default 0.0.</div>
<dl><dt><span class="strong">Parameters:</span></dt><dd><code>regParam</code> - (undocumented)</dd>
<dt><span class="strong">Returns:</span></dt><dd>(undocumented)</dd></dl>
</li>
</ul>
<a name="setConvergenceTol(double)">
<!-- -->
</a>
<ul class="blockList">
<li class="blockList">
<h4>setConvergenceTol</h4>
<pre>public <a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a> setConvergenceTol(double tolerance)</pre>
<div class="block">Set the convergence tolerance. Default 0.001
convergenceTol is a condition which decides iteration termination.
The end of iteration is decided based on below logic.
- If the norm of the new solution vector is >1, the diff of solution vectors
is compared to relative tolerance which means normalizing by the norm of
the new solution vector.
- If the norm of the new solution vector is <=1, the diff of solution vectors
is compared to absolute tolerance which is not normalizing.
Must be between 0.0 and 1.0 inclusively.</div>
<dl><dt><span class="strong">Parameters:</span></dt><dd><code>tolerance</code> - (undocumented)</dd>
<dt><span class="strong">Returns:</span></dt><dd>(undocumented)</dd></dl>
</li>
</ul>
<a name="setGradient(org.apache.spark.mllib.optimization.Gradient)">
<!-- -->
</a>
<ul class="blockList">
<li class="blockList">
<h4>setGradient</h4>
<pre>public <a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a> setGradient(<a href="../../../../../org/apache/spark/mllib/optimization/Gradient.html" title="class in org.apache.spark.mllib.optimization">Gradient</a> gradient)</pre>
<div class="block">Set the gradient function (of the loss function of one single data example)
to be used for SGD.</div>
<dl><dt><span class="strong">Parameters:</span></dt><dd><code>gradient</code> - (undocumented)</dd>
<dt><span class="strong">Returns:</span></dt><dd>(undocumented)</dd></dl>
</li>
</ul>
<a name="setUpdater(org.apache.spark.mllib.optimization.Updater)">
<!-- -->
</a>
<ul class="blockList">
<li class="blockList">
<h4>setUpdater</h4>
<pre>public <a href="../../../../../org/apache/spark/mllib/optimization/GradientDescent.html" title="class in org.apache.spark.mllib.optimization">GradientDescent</a> setUpdater(<a href="../../../../../org/apache/spark/mllib/optimization/Updater.html" title="class in org.apache.spark.mllib.optimization">Updater</a> updater)</pre>
<div class="block">Set the updater function to actually perform a gradient step in a given direction.
The updater is responsible to perform the update from the regularization term as well,
and therefore determines what kind or regularization is used, if any.</div>
<dl><dt><span class="strong">Parameters:</span></dt><dd><code>updater</code> - (undocumented)</dd>
<dt><span class="strong">Returns:</span></dt><dd>(undocumented)</dd></dl>
</li>
</ul>
<a name="optimize(org.apache.spark.rdd.RDD, org.apache.spark.mllib.linalg.Vector)">
<!-- -->
</a>
<ul class="blockListLast">
<li class="blockList">
<h4>optimize</h4>
<pre>public <a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a> optimize(<a href="../../../../../org/apache/spark/rdd/RDD.html" title="class in org.apache.spark.rdd">RDD</a><scala.Tuple2<java.lang.Object,<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a>>> data,
<a href="../../../../../org/apache/spark/mllib/linalg/Vector.html" title="interface in org.apache.spark.mllib.linalg">Vector</a> initialWeights)</pre>
<div class="block">:: DeveloperApi ::
Runs gradient descent on the given training data.</div>
<dl>
<dt><strong>Specified by:</strong></dt>
<dd><code><a href="../../../../../org/apache/spark/mllib/optimization/Optimizer.html#optimize(org.apache.spark.rdd.RDD, org.apache.spark.mllib.linalg.Vector)">optimize</a></code> in interface <code><a href="../../../../../org/apache/spark/mllib/optimization/Optimizer.html" title="interface in org.apache.spark.mllib.optimization">Optimizer</a></code></dd>
<dt><span class="strong">Parameters:</span></dt><dd><code>data</code> - training data</dd><dd><code>initialWeights</code> - initial weights</dd>
<dt><span class="strong">Returns:</span></dt><dd>solution vector</dd></dl>
</li>
</ul>
</li>
</ul>
</li>
</ul>
</div>
</div>
<!-- ========= END OF CLASS DATA ========= -->
<!-- ======= START OF BOTTOM NAVBAR ====== -->
<div class="bottomNav"><a name="navbar_bottom">
<!-- -->
</a><a href="#skip-navbar_bottom" title="Skip navigation links"></a><a name="navbar_bottom_firstrow">
<!-- -->
</a>
<ul class="navList" title="Navigation">
<li><a href="../../../../../overview-summary.html">Overview</a></li>
<li><a href="package-summary.html">Package</a></li>
<li class="navBarCell1Rev">Class</li>
<li><a href="package-tree.html">Tree</a></li>
<li><a href="../../../../../deprecated-list.html">Deprecated</a></li>
<li><a href="../../../../../index-all.html">Index</a></li>
<li><a href="../../../../../help-doc.html">Help</a></li>
</ul>
</div>
<div class="subNav">
<ul class="navList">
<li><a href="../../../../../org/apache/spark/mllib/optimization/Gradient.html" title="class in org.apache.spark.mllib.optimization"><span class="strong">Prev Class</span></a></li>
<li><a href="../../../../../org/apache/spark/mllib/optimization/HingeGradient.html" title="class in org.apache.spark.mllib.optimization"><span class="strong">Next Class</span></a></li>
</ul>
<ul class="navList">
<li><a href="../../../../../index.html?org/apache/spark/mllib/optimization/GradientDescent.html" target="_top">Frames</a></li>
<li><a href="GradientDescent.html" target="_top">No Frames</a></li>
</ul>
<ul class="navList" id="allclasses_navbar_bottom">
<li><a href="../../../../../allclasses-noframe.html">All Classes</a></li>
</ul>
<div>
<script type="text/javascript"><!--
allClassesLink = document.getElementById("allclasses_navbar_bottom");
if(window==top) {
allClassesLink.style.display = "block";
}
else {
allClassesLink.style.display = "none";
}
//-->
</script>
</div>
<div>
<ul class="subNavList">
<li>Summary: </li>
<li>Nested | </li>
<li>Field | </li>
<li>Constr | </li>
<li><a href="#method_summary">Method</a></li>
</ul>
<ul class="subNavList">
<li>Detail: </li>
<li>Field | </li>
<li>Constr | </li>
<li><a href="#method_detail">Method</a></li>
</ul>
</div>
<a name="skip-navbar_bottom">
<!-- -->
</a></div>
<!-- ======== END OF BOTTOM NAVBAR ======= -->
<script defer="defer" type="text/javascript" src="../../../../../lib/jquery.js"></script><script defer="defer" type="text/javascript" src="../../../../../lib/api-javadocs.js"></script></body>
</html>
|
