-
Notifications
You must be signed in to change notification settings - Fork 81
/
Copy pathplot_calibration.html
457 lines (407 loc) · 43.4 KB
/
plot_calibration.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
<!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="Probability calibration of classifiers" />
<meta property="og:type" content="website" />
<meta property="og:url" content="https://scikit-learn/stable/auto_examples/calibration/plot_calibration.html" />
<meta property="og:site_name" content="scikit-learn" />
<meta property="og:description" content="When performing classification you often want to predict not only the class label, but also the associated probability. This probability gives you some kind of confidence on the prediction. However..." />
<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="When performing classification you often want to predict not only the class label, but also the associated probability. This probability gives you some kind of confidence on the prediction. However..." />
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Probability calibration of classifiers — scikit-learn 1.2.2 documentation</title>
<link rel="canonical" href="http://scikit-learn.org/stable/auto_examples/calibration/plot_calibration.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="plot_calibration_multiclass.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="Probability Calibration for 3-class classification">Prev</a><a href="index.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="Calibration">Up</a>
<a href="../classification/index.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="Classification">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="#">Probability calibration of classifiers</a><ul>
<li><a class="reference internal" href="#generate-synthetic-dataset">Generate synthetic dataset</a></li>
<li><a class="reference internal" href="#gaussian-naive-bayes">Gaussian Naive-Bayes</a></li>
<li><a class="reference internal" href="#plot-data-and-the-predicted-probabilities">Plot data and the predicted probabilities</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-calibration-plot-calibration-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="probability-calibration-of-classifiers">
<span id="sphx-glr-auto-examples-calibration-plot-calibration-py"></span><h1>Probability calibration of classifiers<a class="headerlink" href="#probability-calibration-of-classifiers" title="Permalink to this heading">¶</a></h1>
<p>When performing classification you often want to predict not only
the class label, but also the associated probability. This probability
gives you some kind of confidence on the prediction. However, not all
classifiers provide well-calibrated probabilities, some being over-confident
while others being under-confident. Thus, a separate calibration of predicted
probabilities is often desirable as a postprocessing. This example illustrates
two different methods for this calibration and evaluates the quality of the
returned probabilities using Brier’s score
(see <a class="reference external" href="https://en.wikipedia.org/wiki/Brier_score">https://en.wikipedia.org/wiki/Brier_score</a>).</p>
<p>Compared are the estimated probability using a Gaussian naive Bayes classifier
without calibration, with a sigmoid calibration, and with a non-parametric
isotonic calibration. One can observe that only the non-parametric model is
able to provide a probability calibration that returns probabilities close
to the expected 0.5 for most of the samples belonging to the middle
cluster with heterogeneous labels. This results in a significantly improved
Brier score.</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># Authors:</span>
<span class="c1"># Mathieu Blondel <[email protected]></span>
<span class="c1"># Alexandre Gramfort <[email protected]></span>
<span class="c1"># Balazs Kegl <[email protected]></span>
<span class="c1"># Jan Hendrik Metzen <[email protected]></span>
<span class="c1"># License: BSD Style.</span>
</pre></div>
</div>
<section id="generate-synthetic-dataset">
<h2>Generate synthetic dataset<a class="headerlink" href="#generate-synthetic-dataset" title="Permalink to this heading">¶</a></h2>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
<span class="kn">from</span> <span class="nn">sklearn.datasets</span> <span class="kn">import</span> <a href="../../modules/generated/sklearn.datasets.make_blobs.html#sklearn.datasets.make_blobs" title="sklearn.datasets.make_blobs" class="sphx-glr-backref-module-sklearn-datasets sphx-glr-backref-type-py-function"><span class="n">make_blobs</span></a>
<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="n">n_samples</span> <span class="o">=</span> <span class="mi">50000</span>
<span class="n">n_bins</span> <span class="o">=</span> <span class="mi">3</span> <span class="c1"># use 3 bins for calibration_curve as we have 3 clusters here</span>
<span class="c1"># Generate 3 blobs with 2 classes where the second blob contains</span>
<span class="c1"># half positive samples and half negative samples. Probability in this</span>
<span class="c1"># blob is therefore 0.5.</span>
<span class="n">centers</span> <span class="o">=</span> <span class="p">[(</span><span class="o">-</span><span class="mi">5</span><span class="p">,</span> <span class="o">-</span><span class="mi">5</span><span class="p">),</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">),</span> <span class="p">(</span><span class="mi">5</span><span class="p">,</span> <span class="mi">5</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.make_blobs.html#sklearn.datasets.make_blobs" title="sklearn.datasets.make_blobs" class="sphx-glr-backref-module-sklearn-datasets sphx-glr-backref-type-py-function"><span class="n">make_blobs</span></a><span class="p">(</span><span class="n">n_samples</span><span class="o">=</span><span class="n">n_samples</span><span class="p">,</span> <span class="n">centers</span><span class="o">=</span><span class="n">centers</span><span class="p">,</span> <span class="n">shuffle</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">42</span><span class="p">)</span>
<span class="n">y</span><span class="p">[:</span> <span class="n">n_samples</span> <span class="o">//</span> <span class="mi">2</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
<span class="n">y</span><span class="p">[</span><span class="n">n_samples</span> <span class="o">//</span> <span class="mi">2</span> <span class="p">:]</span> <span class="o">=</span> <span class="mi">1</span>
<span class="n">sample_weight</span> <span class="o">=</span> <a href="https://numpy.org/doc/stable/reference/random/legacy.html#numpy.random.RandomState" title="numpy.random.RandomState" class="sphx-glr-backref-module-numpy-random sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">RandomState</span></a><span class="p">(</span><span class="mi">42</span><span class="p">)</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="n">y</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
<span class="c1"># split train, test for calibration</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="p">,</span> <span class="n">sw_train</span><span class="p">,</span> <span class="n">sw_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">sample_weight</span><span class="p">,</span> <span class="n">test_size</span><span class="o">=</span><span class="mf">0.9</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">42</span>
<span class="p">)</span>
</pre></div>
</div>
</section>
<section id="gaussian-naive-bayes">
<h2>Gaussian Naive-Bayes<a class="headerlink" href="#gaussian-naive-bayes" title="Permalink to this heading">¶</a></h2>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">sklearn.calibration</span> <span class="kn">import</span> <a href="../../modules/generated/sklearn.calibration.CalibratedClassifierCV.html#sklearn.calibration.CalibratedClassifierCV" title="sklearn.calibration.CalibratedClassifierCV" class="sphx-glr-backref-module-sklearn-calibration sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">CalibratedClassifierCV</span></a>
<span class="kn">from</span> <span class="nn">sklearn.metrics</span> <span class="kn">import</span> <a href="../../modules/generated/sklearn.metrics.brier_score_loss.html#sklearn.metrics.brier_score_loss" title="sklearn.metrics.brier_score_loss" class="sphx-glr-backref-module-sklearn-metrics sphx-glr-backref-type-py-function"><span class="n">brier_score_loss</span></a>
<span class="kn">from</span> <span class="nn">sklearn.naive_bayes</span> <span class="kn">import</span> <a href="../../modules/generated/sklearn.naive_bayes.GaussianNB.html#sklearn.naive_bayes.GaussianNB" title="sklearn.naive_bayes.GaussianNB" class="sphx-glr-backref-module-sklearn-naive_bayes sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">GaussianNB</span></a>
<span class="c1"># With no calibration</span>
<span class="n">clf</span> <span class="o">=</span> <a href="../../modules/generated/sklearn.naive_bayes.GaussianNB.html#sklearn.naive_bayes.GaussianNB" title="sklearn.naive_bayes.GaussianNB" class="sphx-glr-backref-module-sklearn-naive_bayes sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">GaussianNB</span></a><span class="p">()</span>
<span class="n">clf</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">)</span> <span class="c1"># GaussianNB itself does not support sample-weights</span>
<span class="n">prob_pos_clf</span> <span class="o">=</span> <span class="n">clf</span><span class="o">.</span><span class="n">predict_proba</span><span class="p">(</span><span class="n">X_test</span><span class="p">)[:,</span> <span class="mi">1</span><span class="p">]</span>
<span class="c1"># With isotonic calibration</span>
<span class="n">clf_isotonic</span> <span class="o">=</span> <a href="../../modules/generated/sklearn.calibration.CalibratedClassifierCV.html#sklearn.calibration.CalibratedClassifierCV" title="sklearn.calibration.CalibratedClassifierCV" class="sphx-glr-backref-module-sklearn-calibration sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">CalibratedClassifierCV</span></a><span class="p">(</span><span class="n">clf</span><span class="p">,</span> <span class="n">cv</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">method</span><span class="o">=</span><span class="s2">"isotonic"</span><span class="p">)</span>
<span class="n">clf_isotonic</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">sample_weight</span><span class="o">=</span><span class="n">sw_train</span><span class="p">)</span>
<span class="n">prob_pos_isotonic</span> <span class="o">=</span> <span class="n">clf_isotonic</span><span class="o">.</span><span class="n">predict_proba</span><span class="p">(</span><span class="n">X_test</span><span class="p">)[:,</span> <span class="mi">1</span><span class="p">]</span>
<span class="c1"># With sigmoid calibration</span>
<span class="n">clf_sigmoid</span> <span class="o">=</span> <a href="../../modules/generated/sklearn.calibration.CalibratedClassifierCV.html#sklearn.calibration.CalibratedClassifierCV" title="sklearn.calibration.CalibratedClassifierCV" class="sphx-glr-backref-module-sklearn-calibration sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">CalibratedClassifierCV</span></a><span class="p">(</span><span class="n">clf</span><span class="p">,</span> <span class="n">cv</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">method</span><span class="o">=</span><span class="s2">"sigmoid"</span><span class="p">)</span>
<span class="n">clf_sigmoid</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">sample_weight</span><span class="o">=</span><span class="n">sw_train</span><span class="p">)</span>
<span class="n">prob_pos_sigmoid</span> <span class="o">=</span> <span class="n">clf_sigmoid</span><span class="o">.</span><span class="n">predict_proba</span><span class="p">(</span><span class="n">X_test</span><span class="p">)[:,</span> <span class="mi">1</span><span class="p">]</span>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Brier score losses: (the smaller the better)"</span><span class="p">)</span>
<span class="n">clf_score</span> <span class="o">=</span> <a href="../../modules/generated/sklearn.metrics.brier_score_loss.html#sklearn.metrics.brier_score_loss" title="sklearn.metrics.brier_score_loss" class="sphx-glr-backref-module-sklearn-metrics sphx-glr-backref-type-py-function"><span class="n">brier_score_loss</span></a><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">prob_pos_clf</span><span class="p">,</span> <span class="n">sample_weight</span><span class="o">=</span><span class="n">sw_test</span><span class="p">)</span>
<span class="nb">print</span><span class="p">(</span><span class="s2">"No calibration: </span><span class="si">%1.3f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">clf_score</span><span class="p">)</span>
<span class="n">clf_isotonic_score</span> <span class="o">=</span> <a href="../../modules/generated/sklearn.metrics.brier_score_loss.html#sklearn.metrics.brier_score_loss" title="sklearn.metrics.brier_score_loss" class="sphx-glr-backref-module-sklearn-metrics sphx-glr-backref-type-py-function"><span class="n">brier_score_loss</span></a><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">prob_pos_isotonic</span><span class="p">,</span> <span class="n">sample_weight</span><span class="o">=</span><span class="n">sw_test</span><span class="p">)</span>
<span class="nb">print</span><span class="p">(</span><span class="s2">"With isotonic calibration: </span><span class="si">%1.3f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">clf_isotonic_score</span><span class="p">)</span>
<span class="n">clf_sigmoid_score</span> <span class="o">=</span> <a href="../../modules/generated/sklearn.metrics.brier_score_loss.html#sklearn.metrics.brier_score_loss" title="sklearn.metrics.brier_score_loss" class="sphx-glr-backref-module-sklearn-metrics sphx-glr-backref-type-py-function"><span class="n">brier_score_loss</span></a><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">prob_pos_sigmoid</span><span class="p">,</span> <span class="n">sample_weight</span><span class="o">=</span><span class="n">sw_test</span><span class="p">)</span>
<span class="nb">print</span><span class="p">(</span><span class="s2">"With sigmoid calibration: </span><span class="si">%1.3f</span><span class="s2">"</span> <span class="o">%</span> <span class="n">clf_sigmoid_score</span><span class="p">)</span>
</pre></div>
</div>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Brier score losses: (the smaller the better)
No calibration: 0.104
With isotonic calibration: 0.084
With sigmoid calibration: 0.109
</pre></div>
</div>
</section>
<section id="plot-data-and-the-predicted-probabilities">
<h2>Plot data and the predicted probabilities<a class="headerlink" href="#plot-data-and-the-predicted-probabilities" title="Permalink to this heading">¶</a></h2>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">matplotlib</span> <span class="kn">import</span> <span class="n">cm</span>
<span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</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">y_unique</span> <span class="o">=</span> <a href="https://numpy.org/doc/stable/reference/generated/numpy.unique.html#numpy.unique" title="numpy.unique" class="sphx-glr-backref-module-numpy sphx-glr-backref-type-py-function"><span class="n">np</span><span class="o">.</span><span class="n">unique</span></a><span class="p">(</span><span class="n">y</span><span class="p">)</span>
<span class="n">colors</span> <span class="o">=</span> <span class="n">cm</span><span class="o">.</span><span class="n">rainbow</span><span class="p">(</span><a href="https://numpy.org/doc/stable/reference/generated/numpy.linspace.html#numpy.linspace" title="numpy.linspace" class="sphx-glr-backref-module-numpy sphx-glr-backref-type-py-function"><span class="n">np</span><span class="o">.</span><span class="n">linspace</span></a><span class="p">(</span><span class="mf">0.0</span><span class="p">,</span> <span class="mf">1.0</span><span class="p">,</span> <span class="n">y_unique</span><span class="o">.</span><span class="n">size</span><span class="p">))</span>
<span class="k">for</span> <span class="n">this_y</span><span class="p">,</span> <span class="n">color</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">y_unique</span><span class="p">,</span> <span class="n">colors</span><span class="p">):</span>
<span class="n">this_X</span> <span class="o">=</span> <span class="n">X_train</span><span class="p">[</span><span class="n">y_train</span> <span class="o">==</span> <span class="n">this_y</span><span class="p">]</span>
<span class="n">this_sw</span> <span class="o">=</span> <span class="n">sw_train</span><span class="p">[</span><span class="n">y_train</span> <span class="o">==</span> <span class="n">this_y</span><span class="p">]</span>
<a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.scatter.html#matplotlib.pyplot.scatter" title="matplotlib.pyplot.scatter" 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">scatter</span></a><span class="p">(</span>
<span class="n">this_X</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">],</span>
<span class="n">this_X</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">],</span>
<span class="n">s</span><span class="o">=</span><span class="n">this_sw</span> <span class="o">*</span> <span class="mi">50</span><span class="p">,</span>
<span class="n">c</span><span class="o">=</span><span class="n">color</span><span class="p">[</span><a href="https://numpy.org/doc/stable/reference/constants.html#numpy.newaxis" title="numpy.newaxis" class="sphx-glr-backref-module-numpy sphx-glr-backref-type-py-data"><span class="n">np</span><span class="o">.</span><span class="n">newaxis</span></a><span class="p">,</span> <span class="p">:],</span>
<span class="n">alpha</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
<span class="n">edgecolor</span><span class="o">=</span><span class="s2">"k"</span><span class="p">,</span>
<span class="n">label</span><span class="o">=</span><span class="s2">"Class </span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="n">this_y</span><span class="p">,</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="n">loc</span><span class="o">=</span><span class="s2">"best"</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">"Data"</span><span class="p">)</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">order</span> <span class="o">=</span> <a href="https://numpy.org/doc/stable/reference/generated/numpy.lexsort.html#numpy.lexsort" title="numpy.lexsort" class="sphx-glr-backref-module-numpy sphx-glr-backref-type-py-function"><span class="n">np</span><span class="o">.</span><span class="n">lexsort</span></a><span class="p">((</span><span class="n">prob_pos_clf</span><span class="p">,))</span>
<a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.plot.html#matplotlib.pyplot.plot" title="matplotlib.pyplot.plot" 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">plot</span></a><span class="p">(</span><span class="n">prob_pos_clf</span><span class="p">[</span><span class="n">order</span><span class="p">],</span> <span class="s2">"r"</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="s2">"No calibration (</span><span class="si">%1.3f</span><span class="s2">)"</span> <span class="o">%</span> <span class="n">clf_score</span><span class="p">)</span>
<a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.plot.html#matplotlib.pyplot.plot" title="matplotlib.pyplot.plot" 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">plot</span></a><span class="p">(</span>
<span class="n">prob_pos_isotonic</span><span class="p">[</span><span class="n">order</span><span class="p">],</span>
<span class="s2">"g"</span><span class="p">,</span>
<span class="n">linewidth</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span>
<span class="n">label</span><span class="o">=</span><span class="s2">"Isotonic calibration (</span><span class="si">%1.3f</span><span class="s2">)"</span> <span class="o">%</span> <span class="n">clf_isotonic_score</span><span class="p">,</span>
<span class="p">)</span>
<a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.plot.html#matplotlib.pyplot.plot" title="matplotlib.pyplot.plot" 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">plot</span></a><span class="p">(</span>
<span class="n">prob_pos_sigmoid</span><span class="p">[</span><span class="n">order</span><span class="p">],</span>
<span class="s2">"b"</span><span class="p">,</span>
<span class="n">linewidth</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span>
<span class="n">label</span><span class="o">=</span><span class="s2">"Sigmoid calibration (</span><span class="si">%1.3f</span><span class="s2">)"</span> <span class="o">%</span> <span class="n">clf_sigmoid_score</span><span class="p">,</span>
<span class="p">)</span>
<a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.plot.html#matplotlib.pyplot.plot" title="matplotlib.pyplot.plot" 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">plot</span></a><span class="p">(</span>
<a href="https://numpy.org/doc/stable/reference/generated/numpy.linspace.html#numpy.linspace" title="numpy.linspace" class="sphx-glr-backref-module-numpy sphx-glr-backref-type-py-function"><span class="n">np</span><span class="o">.</span><span class="n">linspace</span></a><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">y_test</span><span class="o">.</span><span class="n">size</span><span class="p">,</span> <span class="mi">51</span><span class="p">)[</span><span class="mi">1</span><span class="p">::</span><span class="mi">2</span><span class="p">],</span>
<span class="n">y_test</span><span class="p">[</span><span class="n">order</span><span class="p">]</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">25</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
<span class="s2">"k"</span><span class="p">,</span>
<span class="n">linewidth</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span>
<span class="n">label</span><span class="o">=</span><span class="sa">r</span><span class="s2">"Empirical"</span><span class="p">,</span>
<span class="p">)</span>
<a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.ylim.html#matplotlib.pyplot.ylim" title="matplotlib.pyplot.ylim" 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">ylim</span></a><span class="p">([</span><span class="o">-</span><span class="mf">0.05</span><span class="p">,</span> <span class="mf">1.05</span><span class="p">])</span>
<a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.xlabel.html#matplotlib.pyplot.xlabel" title="matplotlib.pyplot.xlabel" 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">xlabel</span></a><span class="p">(</span><span class="s2">"Instances sorted according to predicted probability (uncalibrated GNB)"</span><span class="p">)</span>
<a href="https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.ylabel.html#matplotlib.pyplot.ylabel" title="matplotlib.pyplot.ylabel" 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">ylabel</span></a><span class="p">(</span><span class="s2">"P(y=1)"</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="n">loc</span><span class="o">=</span><span class="s2">"upper left"</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">"Gaussian naive Bayes probabilities"</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_calibration_001.png" srcset="../../_images/sphx_glr_plot_calibration_001.png" alt="Data" class = "sphx-glr-multi-img"/></li>
<li><img src="../../_images/sphx_glr_plot_calibration_002.png" srcset="../../_images/sphx_glr_plot_calibration_002.png" alt="Gaussian naive Bayes probabilities" class = "sphx-glr-multi-img"/></li>
</ul>
<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes 0.335 seconds)</p>
<div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-auto-examples-calibration-plot-calibration-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/calibration/plot_calibration.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo10.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/0b39f715b5e32f01df3d212b6d822b82/plot_calibration.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_calibration.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/0c15970ac17183d2bf864a9563081aeb/plot_calibration.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_calibration.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/calibration/plot_calibration.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>