-
Notifications
You must be signed in to change notification settings - Fork 81
/
Copy pathplot_model_complexity_influence.html
636 lines (571 loc) · 59.7 KB
/
plot_model_complexity_influence.html
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
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
<!DOCTYPE html>
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
<head>
<meta charset="utf-8">
<meta name="generator" content="Docutils 0.19: https://docutils.sourceforge.io/" />
<meta property="og:title" content="Model Complexity Influence" />
<meta property="og:type" content="website" />
<meta property="og:url" content="https://scikit-learn/stable/auto_examples/applications/plot_model_complexity_influence.html" />
<meta property="og:site_name" content="scikit-learn" />
<meta property="og:description" content="Demonstrate how model complexity influences both prediction accuracy and computational performance. We will be using two datasets:,- Diabetes dataset for regression. This dataset consists of 10 mea..." />
<meta property="og:image" content="https://scikit-learn.org/stable/_static/scikit-learn-logo-small.png" />
<meta property="og:image:alt" content="scikit-learn" />
<meta name="description" content="Demonstrate how model complexity influences both prediction accuracy and computational performance. We will be using two datasets:,- Diabetes dataset for regression. This dataset consists of 10 mea..." />
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Model Complexity Influence — scikit-learn 1.2.2 documentation</title>
<link rel="canonical" href="http://scikit-learn.org/stable/auto_examples/applications/plot_model_complexity_influence.html" />
<link rel="shortcut icon" href="../../_static/favicon.ico"/>
<link rel="stylesheet" href="../../_static/css/vendor/bootstrap.min.css" type="text/css" />
<link rel="stylesheet" href="../../_static/pygments.css" type="text/css" />
<link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
<link rel="stylesheet" href="../../_static/plot_directive.css" type="text/css" />
<link rel="stylesheet" href="../../_static/sg_gallery.css" type="text/css" />
<link rel="stylesheet" href="../../_static/sg_gallery-binder.css" type="text/css" />
<link rel="stylesheet" href="../../_static/sg_gallery-dataframe.css" type="text/css" />
<link rel="stylesheet" href="../../_static/sg_gallery-rendered-html.css" type="text/css" />
<link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
<script id="documentation_options" data-url_root="../../" src="../../_static/documentation_options.js"></script>
<script src="../../_static/js/vendor/jquery-3.6.3.slim.min.js"></script>
</head>
<body>
<nav id="navbar" class="sk-docs-navbar navbar navbar-expand-md navbar-light bg-light py-0">
<div class="container-fluid sk-docs-container px-0">
<a class="navbar-brand py-0" href="../../index.html">
<img
class="sk-brand-img"
src="../../_static/scikit-learn-logo-small.png"
alt="logo"/>
</a>
<button
id="sk-navbar-toggler"
class="navbar-toggler"
type="button"
data-toggle="collapse"
data-target="#navbarSupportedContent"
aria-controls="navbarSupportedContent"
aria-expanded="false"
aria-label="Toggle navigation"
>
<span class="navbar-toggler-icon"></span>
</button>
<div class="sk-navbar-collapse collapse navbar-collapse" id="navbarSupportedContent">
<ul class="navbar-nav mr-auto">
<li class="nav-item">
<a class="sk-nav-link nav-link" href="../../install.html">Install</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" href="../../user_guide.html">User Guide</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" href="../../modules/classes.html">API</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" href="../index.html">Examples</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" target="_blank" rel="noopener noreferrer" href="https://blog.scikit-learn.org/">Community</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../../getting_started.html" >Getting Started</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../../tutorial/index.html" >Tutorial</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../../whats_new/v1.2.html" >What's new</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../../glossary.html" >Glossary</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="https://scikit-learn.org/dev/developers/index.html" target="_blank" rel="noopener noreferrer">Development</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../../faq.html" >FAQ</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../../support.html" >Support</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../../related_projects.html" >Related packages</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../../roadmap.html" >Roadmap</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../../governance.html" >Governance</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../../about.html" >About us</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="https://github.com/scikit-learn/scikit-learn" >GitHub</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="https://scikit-learn.org/dev/versions.html" >Other Versions and Download</a>
</li>
<li class="nav-item dropdown nav-more-item-dropdown">
<a class="sk-nav-link nav-link dropdown-toggle" href="#" id="navbarDropdown" role="button" data-toggle="dropdown" aria-haspopup="true" aria-expanded="false">More</a>
<div class="dropdown-menu" aria-labelledby="navbarDropdown">
<a class="sk-nav-dropdown-item dropdown-item" href="../../getting_started.html" >Getting Started</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../../tutorial/index.html" >Tutorial</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../../whats_new/v1.2.html" >What's new</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../../glossary.html" >Glossary</a>
<a class="sk-nav-dropdown-item dropdown-item" href="https://scikit-learn.org/dev/developers/index.html" target="_blank" rel="noopener noreferrer">Development</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../../faq.html" >FAQ</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../../support.html" >Support</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../../related_projects.html" >Related packages</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../../roadmap.html" >Roadmap</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../../governance.html" >Governance</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../../about.html" >About us</a>
<a class="sk-nav-dropdown-item dropdown-item" href="https://github.com/scikit-learn/scikit-learn" >GitHub</a>
<a class="sk-nav-dropdown-item dropdown-item" href="https://scikit-learn.org/dev/versions.html" >Other Versions and Download</a>
</div>
</li>
</ul>
<div id="searchbox" role="search">
<div class="searchformwrapper">
<form class="search" action="../../search.html" method="get">
<input class="sk-search-text-input" type="text" name="q" aria-labelledby="searchlabel" />
<input class="sk-search-text-btn" type="submit" value="Go" />
</form>
</div>
</div>
</div>
</div>
</nav>
<div class="d-flex" id="sk-doc-wrapper">
<input type="checkbox" name="sk-toggle-checkbox" id="sk-toggle-checkbox">
<label id="sk-sidemenu-toggle" class="sk-btn-toggle-toc btn sk-btn-primary" for="sk-toggle-checkbox">Toggle Menu</label>
<div id="sk-sidebar-wrapper" class="border-right">
<div class="sk-sidebar-toc-wrapper">
<div class="btn-group w-100 mb-2" role="group" aria-label="rellinks">
<a href="svm_gui.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="Libsvm GUI">Prev</a><a href="index.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="Examples based on real world datasets">Up</a>
<a href="plot_out_of_core_classification.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="Out-of-core classification of text documents">Next</a>
</div>
<div class="alert alert-danger p-1 mb-2" role="alert">
<p class="text-center mb-0">
<strong>scikit-learn 1.2.2</strong><br/>
<a href="http://scikit-learn.org/dev/versions.html">Other versions</a>
</p>
</div>
<div class="alert alert-warning p-1 mb-2" role="alert">
<p class="text-center mb-0">
Please <a class="font-weight-bold" href="../../about.html#citing-scikit-learn"><string>cite us</string></a> if you use the software.
</p>
</div>
<div class="sk-sidebar-toc">
<ul>
<li><a class="reference internal" href="#">Model Complexity Influence</a><ul>
<li><a class="reference internal" href="#load-the-data">Load the data</a></li>
<li><a class="reference internal" href="#benchmark-influence">Benchmark influence</a></li>
<li><a class="reference internal" href="#choose-parameters">Choose parameters</a></li>
<li><a class="reference internal" href="#run-the-code-and-plot-the-results">Run the code and plot the results</a></li>
<li><a class="reference internal" href="#conclusion">Conclusion</a></li>
</ul>
</li>
</ul>
</div>
</div>
</div>
<div id="sk-page-content-wrapper">
<div class="sk-page-content container-fluid body px-md-3" role="main">
<div class="sphx-glr-download-link-note admonition note">
<p class="admonition-title">Note</p>
<p>Click <a class="reference internal" href="#sphx-glr-download-auto-examples-applications-plot-model-complexity-influence-py"><span class="std std-ref">here</span></a>
to download the full example code or to run this example in your browser via Binder</p>
</div>
<section class="sphx-glr-example-title" id="model-complexity-influence">
<span id="sphx-glr-auto-examples-applications-plot-model-complexity-influence-py"></span><h1>Model Complexity Influence<a class="headerlink" href="#model-complexity-influence" title="Permalink to this heading">¶</a></h1>
<p>Demonstrate how model complexity influences both prediction accuracy and
computational performance.</p>
<dl class="simple">
<dt>We will be using two datasets:</dt><dd><ul class="simple">
<li><p><a class="reference internal" href="../../datasets/toy_dataset.html#diabetes-dataset"><span class="std std-ref">Diabetes dataset</span></a> for regression.
This dataset consists of 10 measurements taken from diabetes patients.
The task is to predict disease progression;</p></li>
<li><p><a class="reference internal" href="../../datasets/real_world.html#newsgroups-dataset"><span class="std std-ref">The 20 newsgroups text dataset</span></a> for classification. This dataset consists of
newsgroup posts. The task is to predict on which topic (out of 20 topics)
the post is written about.</p></li>
</ul>
</dd>
<dt>We will model the complexity influence on three different estimators:</dt><dd><ul class="simple">
<li><p><a class="reference internal" href="../../modules/generated/sklearn.linear_model.SGDClassifier.html#sklearn.linear_model.SGDClassifier" title="sklearn.linear_model.SGDClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">SGDClassifier</span></code></a> (for classification data)
which implements stochastic gradient descent learning;</p></li>
<li><p><a class="reference internal" href="../../modules/generated/sklearn.svm.NuSVR.html#sklearn.svm.NuSVR" title="sklearn.svm.NuSVR"><code class="xref py py-class docutils literal notranslate"><span class="pre">NuSVR</span></code></a> (for regression data) which implements
Nu support vector regression;</p></li>
<li><p><a class="reference internal" href="../../modules/generated/sklearn.ensemble.GradientBoostingRegressor.html#sklearn.ensemble.GradientBoostingRegressor" title="sklearn.ensemble.GradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">GradientBoostingRegressor</span></code></a> builds an additive
model in a forward stage-wise fashion. Notice that
<a class="reference internal" href="../../modules/generated/sklearn.ensemble.HistGradientBoostingRegressor.html#sklearn.ensemble.HistGradientBoostingRegressor" title="sklearn.ensemble.HistGradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">HistGradientBoostingRegressor</span></code></a> is much faster
than <a class="reference internal" href="../../modules/generated/sklearn.ensemble.GradientBoostingRegressor.html#sklearn.ensemble.GradientBoostingRegressor" title="sklearn.ensemble.GradientBoostingRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">GradientBoostingRegressor</span></code></a> starting with
intermediate datasets (<code class="docutils literal notranslate"><span class="pre">n_samples</span> <span class="pre">>=</span> <span class="pre">10_000</span></code>), which is not the case for
this example.</p></li>
</ul>
</dd>
</dl>
<p>We make the model complexity vary through the choice of relevant model
parameters in each of our selected models. Next, we will measure the influence
on both computational performance (latency) and predictive power (MSE or
Hamming Loss).</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># Authors: Eustache Diemert <[email protected]></span>
<span class="c1"># Maria Telenczuk <https://github.com/maikia></span>
<span class="c1"># Guillaume Lemaitre <[email protected]></span>
<span class="c1"># License: BSD 3 clause</span>
<span class="kn">import</span> <span class="nn">time</span>
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
<span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
<span class="kn">from</span> <span class="nn">sklearn</span> <span class="kn">import</span> <span class="n">datasets</span>
<span class="kn">from</span> <span class="nn">sklearn.model_selection</span> <span class="kn">import</span> <a href="../../modules/generated/sklearn.model_selection.train_test_split.html#sklearn.model_selection.train_test_split" title="sklearn.model_selection.train_test_split" class="sphx-glr-backref-module-sklearn-model_selection sphx-glr-backref-type-py-function"><span class="n">train_test_split</span></a>
<span class="kn">from</span> <span class="nn">sklearn.metrics</span> <span class="kn">import</span> <a href="../../modules/generated/sklearn.metrics.mean_squared_error.html#sklearn.metrics.mean_squared_error" title="sklearn.metrics.mean_squared_error" class="sphx-glr-backref-module-sklearn-metrics sphx-glr-backref-type-py-function"><span class="n">mean_squared_error</span></a>
<span class="kn">from</span> <span class="nn">sklearn.svm</span> <span class="kn">import</span> <a href="../../modules/generated/sklearn.svm.NuSVR.html#sklearn.svm.NuSVR" title="sklearn.svm.NuSVR" class="sphx-glr-backref-module-sklearn-svm sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">NuSVR</span></a>
<span class="kn">from</span> <span class="nn">sklearn.ensemble</span> <span class="kn">import</span> <a href="../../modules/generated/sklearn.ensemble.GradientBoostingRegressor.html#sklearn.ensemble.GradientBoostingRegressor" title="sklearn.ensemble.GradientBoostingRegressor" class="sphx-glr-backref-module-sklearn-ensemble sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">GradientBoostingRegressor</span></a>
<span class="kn">from</span> <span class="nn">sklearn.linear_model</span> <span class="kn">import</span> <a href="../../modules/generated/sklearn.linear_model.SGDClassifier.html#sklearn.linear_model.SGDClassifier" title="sklearn.linear_model.SGDClassifier" class="sphx-glr-backref-module-sklearn-linear_model sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">SGDClassifier</span></a>
<span class="kn">from</span> <span class="nn">sklearn.metrics</span> <span class="kn">import</span> <a href="../../modules/generated/sklearn.metrics.hamming_loss.html#sklearn.metrics.hamming_loss" title="sklearn.metrics.hamming_loss" class="sphx-glr-backref-module-sklearn-metrics sphx-glr-backref-type-py-function"><span class="n">hamming_loss</span></a>
<span class="c1"># Initialize random generator</span>
<a href="https://numpy.org/doc/stable/reference/random/generated/numpy.random.seed.html#numpy.random.seed" title="numpy.random.seed" class="sphx-glr-backref-module-numpy-random sphx-glr-backref-type-py-function"><span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">seed</span></a><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
</pre></div>
</div>
<section id="load-the-data">
<h2>Load the data<a class="headerlink" href="#load-the-data" title="Permalink to this heading">¶</a></h2>
<p>First we load both datasets.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>We are using
<a class="reference internal" href="../../modules/generated/sklearn.datasets.fetch_20newsgroups_vectorized.html#sklearn.datasets.fetch_20newsgroups_vectorized" title="sklearn.datasets.fetch_20newsgroups_vectorized"><code class="xref py py-func docutils literal notranslate"><span class="pre">fetch_20newsgroups_vectorized</span></code></a> to download 20
newsgroups dataset. It returns ready-to-use features.</p>
</div>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p><code class="docutils literal notranslate"><span class="pre">X</span></code> of the 20 newsgroups dataset is a sparse matrix while <code class="docutils literal notranslate"><span class="pre">X</span></code>
of diabetes dataset is a numpy array.</p>
</div>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">generate_data</span><span class="p">(</span><span class="n">case</span><span class="p">):</span>
<span class="w"> </span><span class="sd">"""Generate regression/classification data."""</span>
<span class="k">if</span> <span class="n">case</span> <span class="o">==</span> <span class="s2">"regression"</span><span class="p">:</span>
<span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <a href="../../modules/generated/sklearn.datasets.load_diabetes.html#sklearn.datasets.load_diabetes" title="sklearn.datasets.load_diabetes" class="sphx-glr-backref-module-sklearn-datasets sphx-glr-backref-type-py-function"><span class="n">datasets</span><span class="o">.</span><span class="n">load_diabetes</span></a><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
<span class="n">train_size</span> <span class="o">=</span> <span class="mf">0.8</span>
<span class="k">elif</span> <span class="n">case</span> <span class="o">==</span> <span class="s2">"classification"</span><span class="p">:</span>
<span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <a href="../../modules/generated/sklearn.datasets.fetch_20newsgroups_vectorized.html#sklearn.datasets.fetch_20newsgroups_vectorized" title="sklearn.datasets.fetch_20newsgroups_vectorized" class="sphx-glr-backref-module-sklearn-datasets sphx-glr-backref-type-py-function"><span class="n">datasets</span><span class="o">.</span><span class="n">fetch_20newsgroups_vectorized</span></a><span class="p">(</span><span class="n">subset</span><span class="o">=</span><span class="s2">"all"</span><span class="p">,</span> <span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
<span class="n">train_size</span> <span class="o">=</span> <span class="mf">0.4</span> <span class="c1"># to make the example run faster</span>
<span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> <a href="../../modules/generated/sklearn.model_selection.train_test_split.html#sklearn.model_selection.train_test_split" title="sklearn.model_selection.train_test_split" class="sphx-glr-backref-module-sklearn-model_selection sphx-glr-backref-type-py-function"><span class="n">train_test_split</span></a><span class="p">(</span>
<span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">train_size</span><span class="o">=</span><span class="n">train_size</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">0</span>
<span class="p">)</span>
<span class="n">data</span> <span class="o">=</span> <span class="p">{</span><span class="s2">"X_train"</span><span class="p">:</span> <span class="n">X_train</span><span class="p">,</span> <span class="s2">"X_test"</span><span class="p">:</span> <span class="n">X_test</span><span class="p">,</span> <span class="s2">"y_train"</span><span class="p">:</span> <span class="n">y_train</span><span class="p">,</span> <span class="s2">"y_test"</span><span class="p">:</span> <span class="n">y_test</span><span class="p">}</span>
<span class="k">return</span> <span class="n">data</span>
<span class="n">regression_data</span> <span class="o">=</span> <span class="n">generate_data</span><span class="p">(</span><span class="s2">"regression"</span><span class="p">)</span>
<span class="n">classification_data</span> <span class="o">=</span> <span class="n">generate_data</span><span class="p">(</span><span class="s2">"classification"</span><span class="p">)</span>
</pre></div>
</div>
</section>
<section id="benchmark-influence">
<h2>Benchmark influence<a class="headerlink" href="#benchmark-influence" title="Permalink to this heading">¶</a></h2>
<p>Next, we can calculate the influence of the parameters on the given
estimator. In each round, we will set the estimator with the new value of
<code class="docutils literal notranslate"><span class="pre">changing_param</span></code> and we will be collecting the prediction times, prediction
performance and complexities to see how those changes affect the estimator.
We will calculate the complexity using <code class="docutils literal notranslate"><span class="pre">complexity_computer</span></code> passed as a
parameter.</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">benchmark_influence</span><span class="p">(</span><span class="n">conf</span><span class="p">):</span>
<span class="w"> </span><span class="sd">"""</span>
<span class="sd"> Benchmark influence of `changing_param` on both MSE and latency.</span>
<span class="sd"> """</span>
<span class="n">prediction_times</span> <span class="o">=</span> <span class="p">[]</span>
<span class="n">prediction_powers</span> <span class="o">=</span> <span class="p">[]</span>
<span class="n">complexities</span> <span class="o">=</span> <span class="p">[]</span>
<span class="k">for</span> <span class="n">param_value</span> <span class="ow">in</span> <span class="n">conf</span><span class="p">[</span><span class="s2">"changing_param_values"</span><span class="p">]:</span>
<span class="n">conf</span><span class="p">[</span><span class="s2">"tuned_params"</span><span class="p">][</span><span class="n">conf</span><span class="p">[</span><span class="s2">"changing_param"</span><span class="p">]]</span> <span class="o">=</span> <span class="n">param_value</span>
<span class="n">estimator</span> <span class="o">=</span> <span class="n">conf</span><span class="p">[</span><span class="s2">"estimator"</span><span class="p">](</span><span class="o">**</span><span class="n">conf</span><span class="p">[</span><span class="s2">"tuned_params"</span><span class="p">])</span>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Benchmarking </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="n">estimator</span><span class="p">)</span>
<span class="n">estimator</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">conf</span><span class="p">[</span><span class="s2">"data"</span><span class="p">][</span><span class="s2">"X_train"</span><span class="p">],</span> <span class="n">conf</span><span class="p">[</span><span class="s2">"data"</span><span class="p">][</span><span class="s2">"y_train"</span><span class="p">])</span>
<span class="n">conf</span><span class="p">[</span><span class="s2">"postfit_hook"</span><span class="p">](</span><span class="n">estimator</span><span class="p">)</span>
<span class="n">complexity</span> <span class="o">=</span> <span class="n">conf</span><span class="p">[</span><span class="s2">"complexity_computer"</span><span class="p">](</span><span class="n">estimator</span><span class="p">)</span>
<span class="n">complexities</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">complexity</span><span class="p">)</span>
<span class="n">start_time</span> <span class="o">=</span> <a href="https://docs.python.org/3/library/time.html#time.time" title="time.time" class="sphx-glr-backref-module-time sphx-glr-backref-type-py-function"><span class="n">time</span><span class="o">.</span><span class="n">time</span></a><span class="p">()</span>
<span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">conf</span><span class="p">[</span><span class="s2">"n_samples"</span><span class="p">]):</span>
<span class="n">y_pred</span> <span class="o">=</span> <span class="n">estimator</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">conf</span><span class="p">[</span><span class="s2">"data"</span><span class="p">][</span><span class="s2">"X_test"</span><span class="p">])</span>
<span class="n">elapsed_time</span> <span class="o">=</span> <span class="p">(</span><a href="https://docs.python.org/3/library/time.html#time.time" title="time.time" class="sphx-glr-backref-module-time sphx-glr-backref-type-py-function"><span class="n">time</span><span class="o">.</span><span class="n">time</span></a><span class="p">()</span> <span class="o">-</span> <span class="n">start_time</span><span class="p">)</span> <span class="o">/</span> <span class="nb">float</span><span class="p">(</span><span class="n">conf</span><span class="p">[</span><span class="s2">"n_samples"</span><span class="p">])</span>
<span class="n">prediction_times</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">elapsed_time</span><span class="p">)</span>
<span class="n">pred_score</span> <span class="o">=</span> <span class="n">conf</span><span class="p">[</span><span class="s2">"prediction_performance_computer"</span><span class="p">](</span>
<span class="n">conf</span><span class="p">[</span><span class="s2">"data"</span><span class="p">][</span><span class="s2">"y_test"</span><span class="p">],</span> <span class="n">y_pred</span>
<span class="p">)</span>
<span class="n">prediction_powers</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">pred_score</span><span class="p">)</span>
<span class="nb">print</span><span class="p">(</span>
<span class="s2">"Complexity: </span><span class="si">%d</span><span class="s2"> | </span><span class="si">%s</span><span class="s2">: </span><span class="si">%.4f</span><span class="s2"> | Pred. Time: </span><span class="si">%f</span><span class="s2">s</span><span class="se">\n</span><span class="s2">"</span>
<span class="o">%</span> <span class="p">(</span>
<span class="n">complexity</span><span class="p">,</span>
<span class="n">conf</span><span class="p">[</span><span class="s2">"prediction_performance_label"</span><span class="p">],</span>
<span class="n">pred_score</span><span class="p">,</span>
<span class="n">elapsed_time</span><span class="p">,</span>
<span class="p">)</span>
<span class="p">)</span>
<span class="k">return</span> <span class="n">prediction_powers</span><span class="p">,</span> <span class="n">prediction_times</span><span class="p">,</span> <span class="n">complexities</span>
</pre></div>
</div>
</section>
<section id="choose-parameters">
<h2>Choose parameters<a class="headerlink" href="#choose-parameters" title="Permalink to this heading">¶</a></h2>
<p>We choose the parameters for each of our estimators by making
a dictionary with all the necessary values.
<code class="docutils literal notranslate"><span class="pre">changing_param</span></code> is the name of the parameter which will vary in each
estimator.
Complexity will be defined by the <code class="docutils literal notranslate"><span class="pre">complexity_label</span></code> and calculated using
<code class="docutils literal notranslate"><span class="pre">complexity_computer</span></code>.
Also note that depending on the estimator type we are passing
different data.</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">_count_nonzero_coefficients</span><span class="p">(</span><span class="n">estimator</span><span class="p">):</span>
<span class="n">a</span> <span class="o">=</span> <span class="n">estimator</span><span class="o">.</span><span class="n">coef_</span><span class="o">.</span><span class="n">toarray</span><span class="p">()</span>
<span class="k">return</span> <a href="https://numpy.org/doc/stable/reference/generated/numpy.count_nonzero.html#numpy.count_nonzero" title="numpy.count_nonzero" class="sphx-glr-backref-module-numpy sphx-glr-backref-type-py-function"><span class="n">np</span><span class="o">.</span><span class="n">count_nonzero</span></a><span class="p">(</span><span class="n">a</span><span class="p">)</span>
<span class="n">configurations</span> <span class="o">=</span> <span class="p">[</span>
<span class="p">{</span>
<span class="s2">"estimator"</span><span class="p">:</span> <a href="../../modules/generated/sklearn.linear_model.SGDClassifier.html#sklearn.linear_model.SGDClassifier" title="sklearn.linear_model.SGDClassifier" class="sphx-glr-backref-module-sklearn-linear_model sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">SGDClassifier</span></a><span class="p">,</span>
<span class="s2">"tuned_params"</span><span class="p">:</span> <span class="p">{</span>
<span class="s2">"penalty"</span><span class="p">:</span> <span class="s2">"elasticnet"</span><span class="p">,</span>
<span class="s2">"alpha"</span><span class="p">:</span> <span class="mf">0.001</span><span class="p">,</span>
<span class="s2">"loss"</span><span class="p">:</span> <span class="s2">"modified_huber"</span><span class="p">,</span>
<span class="s2">"fit_intercept"</span><span class="p">:</span> <span class="kc">True</span><span class="p">,</span>
<span class="s2">"tol"</span><span class="p">:</span> <span class="mf">1e-1</span><span class="p">,</span>
<span class="s2">"n_iter_no_change"</span><span class="p">:</span> <span class="mi">2</span><span class="p">,</span>
<span class="p">},</span>
<span class="s2">"changing_param"</span><span class="p">:</span> <span class="s2">"l1_ratio"</span><span class="p">,</span>
<span class="s2">"changing_param_values"</span><span class="p">:</span> <span class="p">[</span><span class="mf">0.25</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.75</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">],</span>
<span class="s2">"complexity_label"</span><span class="p">:</span> <span class="s2">"non_zero coefficients"</span><span class="p">,</span>
<span class="s2">"complexity_computer"</span><span class="p">:</span> <span class="n">_count_nonzero_coefficients</span><span class="p">,</span>
<span class="s2">"prediction_performance_computer"</span><span class="p">:</span> <a href="../../modules/generated/sklearn.metrics.hamming_loss.html#sklearn.metrics.hamming_loss" title="sklearn.metrics.hamming_loss" class="sphx-glr-backref-module-sklearn-metrics sphx-glr-backref-type-py-function"><span class="n">hamming_loss</span></a><span class="p">,</span>
<span class="s2">"prediction_performance_label"</span><span class="p">:</span> <span class="s2">"Hamming Loss (Misclassification Ratio)"</span><span class="p">,</span>
<span class="s2">"postfit_hook"</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="o">.</span><span class="n">sparsify</span><span class="p">(),</span>
<span class="s2">"data"</span><span class="p">:</span> <span class="n">classification_data</span><span class="p">,</span>
<span class="s2">"n_samples"</span><span class="p">:</span> <span class="mi">5</span><span class="p">,</span>
<span class="p">},</span>
<span class="p">{</span>
<span class="s2">"estimator"</span><span class="p">:</span> <a href="../../modules/generated/sklearn.svm.NuSVR.html#sklearn.svm.NuSVR" title="sklearn.svm.NuSVR" class="sphx-glr-backref-module-sklearn-svm sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">NuSVR</span></a><span class="p">,</span>
<span class="s2">"tuned_params"</span><span class="p">:</span> <span class="p">{</span><span class="s2">"C"</span><span class="p">:</span> <span class="mf">1e3</span><span class="p">,</span> <span class="s2">"gamma"</span><span class="p">:</span> <span class="mi">2</span><span class="o">**-</span><span class="mi">15</span><span class="p">},</span>
<span class="s2">"changing_param"</span><span class="p">:</span> <span class="s2">"nu"</span><span class="p">,</span>
<span class="s2">"changing_param_values"</span><span class="p">:</span> <span class="p">[</span><span class="mf">0.05</span><span class="p">,</span> <span class="mf">0.1</span><span class="p">,</span> <span class="mf">0.2</span><span class="p">,</span> <span class="mf">0.35</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">],</span>
<span class="s2">"complexity_label"</span><span class="p">:</span> <span class="s2">"n_support_vectors"</span><span class="p">,</span>
<span class="s2">"complexity_computer"</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">support_vectors_</span><span class="p">),</span>
<span class="s2">"data"</span><span class="p">:</span> <span class="n">regression_data</span><span class="p">,</span>
<span class="s2">"postfit_hook"</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">,</span>
<span class="s2">"prediction_performance_computer"</span><span class="p">:</span> <a href="../../modules/generated/sklearn.metrics.mean_squared_error.html#sklearn.metrics.mean_squared_error" title="sklearn.metrics.mean_squared_error" class="sphx-glr-backref-module-sklearn-metrics sphx-glr-backref-type-py-function"><span class="n">mean_squared_error</span></a><span class="p">,</span>
<span class="s2">"prediction_performance_label"</span><span class="p">:</span> <span class="s2">"MSE"</span><span class="p">,</span>
<span class="s2">"n_samples"</span><span class="p">:</span> <span class="mi">15</span><span class="p">,</span>
<span class="p">},</span>
<span class="p">{</span>
<span class="s2">"estimator"</span><span class="p">:</span> <a href="../../modules/generated/sklearn.ensemble.GradientBoostingRegressor.html#sklearn.ensemble.GradientBoostingRegressor" title="sklearn.ensemble.GradientBoostingRegressor" class="sphx-glr-backref-module-sklearn-ensemble sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">GradientBoostingRegressor</span></a><span class="p">,</span>
<span class="s2">"tuned_params"</span><span class="p">:</span> <span class="p">{</span>
<span class="s2">"loss"</span><span class="p">:</span> <span class="s2">"squared_error"</span><span class="p">,</span>
<span class="s2">"learning_rate"</span><span class="p">:</span> <span class="mf">0.05</span><span class="p">,</span>
<span class="s2">"max_depth"</span><span class="p">:</span> <span class="mi">2</span><span class="p">,</span>
<span class="p">},</span>
<span class="s2">"changing_param"</span><span class="p">:</span> <span class="s2">"n_estimators"</span><span class="p">,</span>
<span class="s2">"changing_param_values"</span><span class="p">:</span> <span class="p">[</span><span class="mi">10</span><span class="p">,</span> <span class="mi">25</span><span class="p">,</span> <span class="mi">50</span><span class="p">,</span> <span class="mi">75</span><span class="p">,</span> <span class="mi">100</span><span class="p">],</span>
<span class="s2">"complexity_label"</span><span class="p">:</span> <span class="s2">"n_trees"</span><span class="p">,</span>
<span class="s2">"complexity_computer"</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="o">.</span><span class="n">n_estimators</span><span class="p">,</span>
<span class="s2">"data"</span><span class="p">:</span> <span class="n">regression_data</span><span class="p">,</span>
<span class="s2">"postfit_hook"</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">,</span>
<span class="s2">"prediction_performance_computer"</span><span class="p">:</span> <a href="../../modules/generated/sklearn.metrics.mean_squared_error.html#sklearn.metrics.mean_squared_error" title="sklearn.metrics.mean_squared_error" class="sphx-glr-backref-module-sklearn-metrics sphx-glr-backref-type-py-function"><span class="n">mean_squared_error</span></a><span class="p">,</span>
<span class="s2">"prediction_performance_label"</span><span class="p">:</span> <span class="s2">"MSE"</span><span class="p">,</span>
<span class="s2">"n_samples"</span><span class="p">:</span> <span class="mi">15</span><span class="p">,</span>
<span class="p">},</span>
<span class="p">]</span>
</pre></div>
</div>
</section>
<section id="run-the-code-and-plot-the-results">
<h2>Run the code and plot the results<a class="headerlink" href="#run-the-code-and-plot-the-results" title="Permalink to this heading">¶</a></h2>
<p>We defined all the functions required to run our benchmark. Now, we will loop
over the different configurations that we defined previously. Subsequently,
we can analyze the plots obtained from the benchmark:
Relaxing the <code class="docutils literal notranslate"><span class="pre">L1</span></code> penalty in the SGD classifier reduces the prediction error
but leads to an increase in the training time.
We can draw a similar analysis regarding the training time which increases
with the number of support vectors with a Nu-SVR. However, we observed that
there is an optimal number of support vectors which reduces the prediction
error. Indeed, too few support vectors lead to an under-fitted model while
too many support vectors lead to an over-fitted model.
The exact same conclusion can be drawn for the gradient-boosting model. The
only the difference with the Nu-SVR is that having too many trees in the
ensemble is not as detrimental.</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">plot_influence</span><span class="p">(</span><span class="n">conf</span><span class="p">,</span> <span class="n">mse_values</span><span class="p">,</span> <span class="n">prediction_times</span><span class="p">,</span> <span class="n">complexities</span><span class="p">):</span>
<span class="w"> </span><span class="sd">"""</span>
<span class="sd"> Plot influence of model complexity on both accuracy and latency.</span>
<span class="sd"> """</span>
<span class="n">fig</span> <span class="o">=</span> <a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.figure.html#matplotlib.pyplot.figure" title="matplotlib.pyplot.figure" class="sphx-glr-backref-module-matplotlib-pyplot sphx-glr-backref-type-py-function"><span class="n">plt</span><span class="o">.</span><span class="n">figure</span></a><span class="p">()</span>
<span class="n">fig</span><span class="o">.</span><span class="n">subplots_adjust</span><span class="p">(</span><span class="n">right</span><span class="o">=</span><span class="mf">0.75</span><span class="p">)</span>
<span class="c1"># first axes (prediction error)</span>
<span class="n">ax1</span> <span class="o">=</span> <span class="n">fig</span><span class="o">.</span><span class="n">add_subplot</span><span class="p">(</span><span class="mi">111</span><span class="p">)</span>
<span class="n">line1</span> <span class="o">=</span> <span class="n">ax1</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span><span class="n">complexities</span><span class="p">,</span> <span class="n">mse_values</span><span class="p">,</span> <span class="n">c</span><span class="o">=</span><span class="s2">"tab:blue"</span><span class="p">,</span> <span class="n">ls</span><span class="o">=</span><span class="s2">"-"</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
<span class="n">ax1</span><span class="o">.</span><span class="n">set_xlabel</span><span class="p">(</span><span class="s2">"Model Complexity (</span><span class="si">%s</span><span class="s2">)"</span> <span class="o">%</span> <span class="n">conf</span><span class="p">[</span><span class="s2">"complexity_label"</span><span class="p">])</span>
<span class="n">y1_label</span> <span class="o">=</span> <span class="n">conf</span><span class="p">[</span><span class="s2">"prediction_performance_label"</span><span class="p">]</span>
<span class="n">ax1</span><span class="o">.</span><span class="n">set_ylabel</span><span class="p">(</span><span class="n">y1_label</span><span class="p">)</span>
<span class="n">ax1</span><span class="o">.</span><span class="n">spines</span><span class="p">[</span><span class="s2">"left"</span><span class="p">]</span><span class="o">.</span><span class="n">set_color</span><span class="p">(</span><span class="n">line1</span><span class="o">.</span><span class="n">get_color</span><span class="p">())</span>
<span class="n">ax1</span><span class="o">.</span><span class="n">yaxis</span><span class="o">.</span><span class="n">label</span><span class="o">.</span><span class="n">set_color</span><span class="p">(</span><span class="n">line1</span><span class="o">.</span><span class="n">get_color</span><span class="p">())</span>
<span class="n">ax1</span><span class="o">.</span><span class="n">tick_params</span><span class="p">(</span><span class="n">axis</span><span class="o">=</span><span class="s2">"y"</span><span class="p">,</span> <span class="n">colors</span><span class="o">=</span><span class="n">line1</span><span class="o">.</span><span class="n">get_color</span><span class="p">())</span>
<span class="c1"># second axes (latency)</span>
<span class="n">ax2</span> <span class="o">=</span> <span class="n">fig</span><span class="o">.</span><span class="n">add_subplot</span><span class="p">(</span><span class="mi">111</span><span class="p">,</span> <span class="n">sharex</span><span class="o">=</span><span class="n">ax1</span><span class="p">,</span> <span class="n">frameon</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
<span class="n">line2</span> <span class="o">=</span> <span class="n">ax2</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span><span class="n">complexities</span><span class="p">,</span> <span class="n">prediction_times</span><span class="p">,</span> <span class="n">c</span><span class="o">=</span><span class="s2">"tab:orange"</span><span class="p">,</span> <span class="n">ls</span><span class="o">=</span><span class="s2">"-"</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
<span class="n">ax2</span><span class="o">.</span><span class="n">yaxis</span><span class="o">.</span><span class="n">tick_right</span><span class="p">()</span>
<span class="n">ax2</span><span class="o">.</span><span class="n">yaxis</span><span class="o">.</span><span class="n">set_label_position</span><span class="p">(</span><span class="s2">"right"</span><span class="p">)</span>
<span class="n">y2_label</span> <span class="o">=</span> <span class="s2">"Time (s)"</span>
<span class="n">ax2</span><span class="o">.</span><span class="n">set_ylabel</span><span class="p">(</span><span class="n">y2_label</span><span class="p">)</span>
<span class="n">ax1</span><span class="o">.</span><span class="n">spines</span><span class="p">[</span><span class="s2">"right"</span><span class="p">]</span><span class="o">.</span><span class="n">set_color</span><span class="p">(</span><span class="n">line2</span><span class="o">.</span><span class="n">get_color</span><span class="p">())</span>
<span class="n">ax2</span><span class="o">.</span><span class="n">yaxis</span><span class="o">.</span><span class="n">label</span><span class="o">.</span><span class="n">set_color</span><span class="p">(</span><span class="n">line2</span><span class="o">.</span><span class="n">get_color</span><span class="p">())</span>
<span class="n">ax2</span><span class="o">.</span><span class="n">tick_params</span><span class="p">(</span><span class="n">axis</span><span class="o">=</span><span class="s2">"y"</span><span class="p">,</span> <span class="n">colors</span><span class="o">=</span><span class="n">line2</span><span class="o">.</span><span class="n">get_color</span><span class="p">())</span>
<a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.legend.html#matplotlib.pyplot.legend" title="matplotlib.pyplot.legend" class="sphx-glr-backref-module-matplotlib-pyplot sphx-glr-backref-type-py-function"><span class="n">plt</span><span class="o">.</span><span class="n">legend</span></a><span class="p">(</span>
<span class="p">(</span><span class="n">line1</span><span class="p">,</span> <span class="n">line2</span><span class="p">),</span> <span class="p">(</span><span class="s2">"prediction error"</span><span class="p">,</span> <span class="s2">"prediction latency"</span><span class="p">),</span> <span class="n">loc</span><span class="o">=</span><span class="s2">"upper center"</span>
<span class="p">)</span>
<a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.title.html#matplotlib.pyplot.title" title="matplotlib.pyplot.title" class="sphx-glr-backref-module-matplotlib-pyplot sphx-glr-backref-type-py-function"><span class="n">plt</span><span class="o">.</span><span class="n">title</span></a><span class="p">(</span>
<span class="s2">"Influence of varying '</span><span class="si">%s</span><span class="s2">' on </span><span class="si">%s</span><span class="s2">"</span>
<span class="o">%</span> <span class="p">(</span><span class="n">conf</span><span class="p">[</span><span class="s2">"changing_param"</span><span class="p">],</span> <span class="n">conf</span><span class="p">[</span><span class="s2">"estimator"</span><span class="p">]</span><span class="o">.</span><span class="vm">__name__</span><span class="p">)</span>
<span class="p">)</span>
<span class="k">for</span> <span class="n">conf</span> <span class="ow">in</span> <span class="n">configurations</span><span class="p">:</span>
<span class="n">prediction_performances</span><span class="p">,</span> <span class="n">prediction_times</span><span class="p">,</span> <span class="n">complexities</span> <span class="o">=</span> <span class="n">benchmark_influence</span><span class="p">(</span><span class="n">conf</span><span class="p">)</span>
<span class="n">plot_influence</span><span class="p">(</span><span class="n">conf</span><span class="p">,</span> <span class="n">prediction_performances</span><span class="p">,</span> <span class="n">prediction_times</span><span class="p">,</span> <span class="n">complexities</span><span class="p">)</span>
<a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.show.html#matplotlib.pyplot.show" title="matplotlib.pyplot.show" class="sphx-glr-backref-module-matplotlib-pyplot sphx-glr-backref-type-py-function"><span class="n">plt</span><span class="o">.</span><span class="n">show</span></a><span class="p">()</span>
</pre></div>
</div>
<ul class="sphx-glr-horizontal">
<li><img src="../../_images/sphx_glr_plot_model_complexity_influence_001.png" srcset="../../_images/sphx_glr_plot_model_complexity_influence_001.png" alt="Influence of varying 'l1_ratio' on SGDClassifier" class = "sphx-glr-multi-img"/></li>
<li><img src="../../_images/sphx_glr_plot_model_complexity_influence_002.png" srcset="../../_images/sphx_glr_plot_model_complexity_influence_002.png" alt="Influence of varying 'nu' on NuSVR" class = "sphx-glr-multi-img"/></li>
<li><img src="../../_images/sphx_glr_plot_model_complexity_influence_003.png" srcset="../../_images/sphx_glr_plot_model_complexity_influence_003.png" alt="Influence of varying 'n_estimators' on GradientBoostingRegressor" class = "sphx-glr-multi-img"/></li>
</ul>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Benchmarking SGDClassifier(alpha=0.001, l1_ratio=0.25, loss='modified_huber',
n_iter_no_change=2, penalty='elasticnet', tol=0.1)
Complexity: 4948 | Hamming Loss (Misclassification Ratio): 0.2675 | Pred. Time: 0.065939s
Benchmarking SGDClassifier(alpha=0.001, l1_ratio=0.5, loss='modified_huber',
n_iter_no_change=2, penalty='elasticnet', tol=0.1)
Complexity: 1847 | Hamming Loss (Misclassification Ratio): 0.3264 | Pred. Time: 0.049992s
Benchmarking SGDClassifier(alpha=0.001, l1_ratio=0.75, loss='modified_huber',
n_iter_no_change=2, penalty='elasticnet', tol=0.1)
Complexity: 997 | Hamming Loss (Misclassification Ratio): 0.3383 | Pred. Time: 0.038358s
Benchmarking SGDClassifier(alpha=0.001, l1_ratio=0.9, loss='modified_huber',
n_iter_no_change=2, penalty='elasticnet', tol=0.1)
Complexity: 802 | Hamming Loss (Misclassification Ratio): 0.3582 | Pred. Time: 0.035313s
Benchmarking NuSVR(C=1000.0, gamma=3.0517578125e-05, nu=0.05)
Complexity: 18 | MSE: 5558.7313 | Pred. Time: 0.000160s
Benchmarking NuSVR(C=1000.0, gamma=3.0517578125e-05, nu=0.1)
Complexity: 36 | MSE: 5289.8022 | Pred. Time: 0.000237s
Benchmarking NuSVR(C=1000.0, gamma=3.0517578125e-05, nu=0.2)
Complexity: 72 | MSE: 5193.8353 | Pred. Time: 0.000394s
Benchmarking NuSVR(C=1000.0, gamma=3.0517578125e-05, nu=0.35)
Complexity: 124 | MSE: 5131.3279 | Pred. Time: 0.000620s
Benchmarking NuSVR(C=1000.0, gamma=3.0517578125e-05)
Complexity: 178 | MSE: 5149.0779 | Pred. Time: 0.000860s
Benchmarking GradientBoostingRegressor(learning_rate=0.05, max_depth=2, n_estimators=10)
Complexity: 10 | MSE: 4066.4812 | Pred. Time: 0.000118s
Benchmarking GradientBoostingRegressor(learning_rate=0.05, max_depth=2, n_estimators=25)
Complexity: 25 | MSE: 3551.1723 | Pred. Time: 0.000132s
Benchmarking GradientBoostingRegressor(learning_rate=0.05, max_depth=2, n_estimators=50)
Complexity: 50 | MSE: 3445.2171 | Pred. Time: 0.000182s
Benchmarking GradientBoostingRegressor(learning_rate=0.05, max_depth=2, n_estimators=75)
Complexity: 75 | MSE: 3433.0358 | Pred. Time: 0.000196s
Benchmarking GradientBoostingRegressor(learning_rate=0.05, max_depth=2)
Complexity: 100 | MSE: 3456.0602 | Pred. Time: 0.000225s
</pre></div>
</div>
</section>
<section id="conclusion">
<h2>Conclusion<a class="headerlink" href="#conclusion" title="Permalink to this heading">¶</a></h2>
<p>As a conclusion, we can deduce the following insights:</p>
<ul class="simple">
<li><p>a model which is more complex (or expressive) will require a larger
training time;</p></li>
<li><p>a more complex model does not guarantee to reduce the prediction error.</p></li>
</ul>
<p>These aspects are related to model generalization and avoiding model
under-fitting or over-fitting.</p>
<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes 18.633 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-applications-plot-model-complexity-influence-py">
<div class="binder-badge docutils container">
<a class="reference external image-reference" href="https://mybinder.org/v2/gh/scikit-learn/scikit-learn/1.2.X?urlpath=lab/tree/notebooks/auto_examples/applications/plot_model_complexity_influence.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo8.svg" width="150px" /></a>
</div>
<div class="sphx-glr-download sphx-glr-download-python docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/ddd79923ba48c7f71fb17697baa1a22b/plot_model_complexity_influence.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">plot_model_complexity_influence.py</span></code></a></p>
</div>
<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
<p><a class="reference download internal" download="" href="../../_downloads/3ed102fa8211c8d36f2331f0c5e1dcef/plot_model_complexity_influence.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">plot_model_complexity_influence.ipynb</span></code></a></p>
</div>
</div>
<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
</section>
</section>
</div>
<div class="container">
<footer class="sk-content-footer">
© 2007 - 2023, scikit-learn developers (BSD License).
<a href="../../_sources/auto_examples/applications/plot_model_complexity_influence.rst.txt" rel="nofollow">Show this page source</a>
</footer>
</div>
</div>
</div>
<script src="../../_static/js/vendor/bootstrap.min.js"></script>
<script>
window.ga=window.ga||function(){(ga.q=ga.q||[]).push(arguments)};ga.l=+new Date;
ga('create', 'UA-22606712-2', 'auto');
ga('set', 'anonymizeIp', true);
ga('send', 'pageview');
</script>
<script async src='https://www.google-analytics.com/analytics.js'></script>
<script defer data-domain="scikit-learn.org" src="https://views.scientific-python.org/js/script.js">
</script>
<script>
$(document).ready(function() {
/* Add a [>>>] button on the top-right corner of code samples to hide
* the >>> and ... prompts and the output and thus make the code
* copyable. */
var div = $('.highlight-python .highlight,' +
'.highlight-python3 .highlight,' +
'.highlight-pycon .highlight,' +
'.highlight-default .highlight')
var pre = div.find('pre');
// get the styles from the current theme
pre.parent().parent().css('position', 'relative');
var hide_text = 'Hide prompts and outputs';
var show_text = 'Show prompts and outputs';
// create and add the button to all the code blocks that contain >>>
div.each(function(index) {
var jthis = $(this);
if (jthis.find('.gp').length > 0) {
var button = $('<span class="copybutton">>>></span>');
button.attr('title', hide_text);
button.data('hidden', 'false');
jthis.prepend(button);
}
// tracebacks (.gt) contain bare text elements that need to be
// wrapped in a span to work with .nextUntil() (see later)
jthis.find('pre:has(.gt)').contents().filter(function() {
return ((this.nodeType == 3) && (this.data.trim().length > 0));
}).wrap('<span>');
});
// define the behavior of the button when it's clicked
$('.copybutton').click(function(e){
e.preventDefault();
var button = $(this);
if (button.data('hidden') === 'false') {
// hide the code output
button.parent().find('.go, .gp, .gt').hide();
button.next('pre').find('.gt').nextUntil('.gp, .go').css('visibility', 'hidden');
button.css('text-decoration', 'line-through');
button.attr('title', show_text);
button.data('hidden', 'true');
} else {
// show the code output
button.parent().find('.go, .gp, .gt').show();
button.next('pre').find('.gt').nextUntil('.gp, .go').css('visibility', 'visible');
button.css('text-decoration', 'none');
button.attr('title', hide_text);
button.data('hidden', 'false');
}
});
/*** Add permalink buttons next to glossary terms ***/
$('dl.glossary > dt[id]').append(function() {
return ('<a class="headerlink" href="#' +
this.getAttribute('id') +
'" title="Permalink to this term">¶</a>');
});
});
</script>
<script id="MathJax-script" async src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-chtml.js"></script>
<script src="https://scikit-learn.org/versionwarning.js"></script>
</body>
</html>