-
Notifications
You must be signed in to change notification settings - Fork 81
/
Copy pathtree.html
862 lines (822 loc) · 71.8 KB
/
tree.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
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
<!DOCTYPE html>
<!-- data-theme below is forced to be "light" but should be changed if we use pydata-theme-sphinx in the future -->
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" data-content_root="../" data-theme="light"> <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" data-content_root="../" data-theme="light"> <!--<![endif]-->
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1" />
<meta property="og:title" content="1.10. Decision Trees" />
<meta property="og:type" content="website" />
<meta property="og:url" content="https://scikit-learn/stable/modules/tree.html" />
<meta property="og:site_name" content="scikit-learn" />
<meta property="og:description" content="Decision Trees (DTs) are a non-parametric supervised learning method used for classification and regression. The goal is to create a model that predicts the value of a target variable by learning s..." />
<meta property="og:image" content="https://scikit-learn/stable/_images/sphx_glr_plot_tree_regression_001.png" />
<meta property="og:image:alt" content="scikit-learn" />
<meta name="description" content="Decision Trees (DTs) are a non-parametric supervised learning method used for classification and regression. The goal is to create a model that predicts the value of a target variable by learning s..." />
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>1.10. Decision Trees — scikit-learn 1.5.dev0 documentation</title>
<link rel="canonical" href="http://scikit-learn.org/stable/modules/tree.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/copybutton.css" type="text/css" />
<link rel="stylesheet" href="../_static/plot_directive.css" type="text/css" />
<link rel="stylesheet" href="../https://fonts.googleapis.com/css?family=Vibur" type="text/css" />
<link rel="stylesheet" href="../_static/jupyterlite_sphinx.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>
<script src="../_static/js/details-permalink.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="classes.html">API</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" href="../auto_examples/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/v1.5.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="../developers/index.html" >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/v1.5.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="../developers/index.html" >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="naive_bayes.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="1.9. Naive Bayes">Prev</a><a href="../supervised_learning.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="1. Supervised learning">Up</a>
<a href="ensemble.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="1.11. Ensembles: Gradient boosting, random forests, bagging, voting, stacking">Next</a>
</div>
<div class="alert alert-danger p-1 mb-2" role="alert">
<p class="text-center mb-0">
<strong>scikit-learn 1.5.dev0</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="#">1.10. Decision Trees</a><ul>
<li><a class="reference internal" href="#classification">1.10.1. Classification</a></li>
<li><a class="reference internal" href="#regression">1.10.2. Regression</a></li>
<li><a class="reference internal" href="#multi-output-problems">1.10.3. Multi-output problems</a></li>
<li><a class="reference internal" href="#complexity">1.10.4. Complexity</a></li>
<li><a class="reference internal" href="#tips-on-practical-use">1.10.5. Tips on practical use</a></li>
<li><a class="reference internal" href="#tree-algorithms-id3-c4-5-c5-0-and-cart">1.10.6. Tree algorithms: ID3, C4.5, C5.0 and CART</a></li>
<li><a class="reference internal" href="#mathematical-formulation">1.10.7. Mathematical formulation</a><ul>
<li><a class="reference internal" href="#classification-criteria">1.10.7.1. Classification criteria</a></li>
<li><a class="reference internal" href="#regression-criteria">1.10.7.2. Regression criteria</a></li>
</ul>
</li>
<li><a class="reference internal" href="#missing-values-support">1.10.8. Missing Values Support</a></li>
<li><a class="reference internal" href="#minimal-cost-complexity-pruning">1.10.9. Minimal Cost-Complexity Pruning</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">
<section id="decision-trees">
<span id="tree"></span><h1><span class="section-number">1.10. </span>Decision Trees<a class="headerlink" href="#decision-trees" title="Link to this heading">¶</a></h1>
<p><strong>Decision Trees (DTs)</strong> are a non-parametric supervised learning method used
for <a class="reference internal" href="#tree-classification"><span class="std std-ref">classification</span></a> and <a class="reference internal" href="#tree-regression"><span class="std std-ref">regression</span></a>. The goal is to create a model that predicts the value of a
target variable by learning simple decision rules inferred from the data
features. A tree can be seen as a piecewise constant approximation.</p>
<p>For instance, in the example below, decision trees learn from data to
approximate a sine curve with a set of if-then-else decision rules. The deeper
the tree, the more complex the decision rules and the fitter the model.</p>
<figure class="align-center">
<a class="reference external image-reference" href="../auto_examples/tree/plot_tree_regression.html"><img alt="../_images/sphx_glr_plot_tree_regression_001.png" src="../_images/sphx_glr_plot_tree_regression_001.png" style="width: 480.0px; height: 360.0px;" /></a>
</figure>
<p>Some advantages of decision trees are:</p>
<ul class="simple">
<li><p>Simple to understand and to interpret. Trees can be visualized.</p></li>
<li><p>Requires little data preparation. Other techniques often require data
normalization, dummy variables need to be created and blank values to
be removed. Some tree and algorithm combinations support
<a class="reference internal" href="#tree-missing-value-support"><span class="std std-ref">missing values</span></a>.</p></li>
<li><p>The cost of using the tree (i.e., predicting data) is logarithmic in the
number of data points used to train the tree.</p></li>
<li><p>Able to handle both numerical and categorical data. However, the scikit-learn
implementation does not support categorical variables for now. Other
techniques are usually specialized in analyzing datasets that have only one type
of variable. See <a class="reference internal" href="#tree-algorithms"><span class="std std-ref">algorithms</span></a> for more
information.</p></li>
<li><p>Able to handle multi-output problems.</p></li>
<li><p>Uses a white box model. If a given situation is observable in a model,
the explanation for the condition is easily explained by boolean logic.
By contrast, in a black box model (e.g., in an artificial neural
network), results may be more difficult to interpret.</p></li>
<li><p>Possible to validate a model using statistical tests. That makes it
possible to account for the reliability of the model.</p></li>
<li><p>Performs well even if its assumptions are somewhat violated by
the true model from which the data were generated.</p></li>
</ul>
<p>The disadvantages of decision trees include:</p>
<ul class="simple">
<li><p>Decision-tree learners can create over-complex trees that do not
generalize the data well. This is called overfitting. Mechanisms
such as pruning, setting the minimum number of samples required
at a leaf node or setting the maximum depth of the tree are
necessary to avoid this problem.</p></li>
<li><p>Decision trees can be unstable because small variations in the
data might result in a completely different tree being generated.
This problem is mitigated by using decision trees within an
ensemble.</p></li>
<li><p>Predictions of decision trees are neither smooth nor continuous, but
piecewise constant approximations as seen in the above figure. Therefore,
they are not good at extrapolation.</p></li>
<li><p>The problem of learning an optimal decision tree is known to be
NP-complete under several aspects of optimality and even for simple
concepts. Consequently, practical decision-tree learning algorithms
are based on heuristic algorithms such as the greedy algorithm where
locally optimal decisions are made at each node. Such algorithms
cannot guarantee to return the globally optimal decision tree. This
can be mitigated by training multiple trees in an ensemble learner,
where the features and samples are randomly sampled with replacement.</p></li>
<li><p>There are concepts that are hard to learn because decision trees
do not express them easily, such as XOR, parity or multiplexer problems.</p></li>
<li><p>Decision tree learners create biased trees if some classes dominate.
It is therefore recommended to balance the dataset prior to fitting
with the decision tree.</p></li>
</ul>
<section id="classification">
<span id="tree-classification"></span><h2><span class="section-number">1.10.1. </span>Classification<a class="headerlink" href="#classification" title="Link to this heading">¶</a></h2>
<p><a class="reference internal" href="generated/sklearn.tree.DecisionTreeClassifier.html#sklearn.tree.DecisionTreeClassifier" title="sklearn.tree.DecisionTreeClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">DecisionTreeClassifier</span></code></a> is a class capable of performing multi-class
classification on a dataset.</p>
<p>As with other classifiers, <a class="reference internal" href="generated/sklearn.tree.DecisionTreeClassifier.html#sklearn.tree.DecisionTreeClassifier" title="sklearn.tree.DecisionTreeClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">DecisionTreeClassifier</span></code></a> takes as input two arrays:
an array X, sparse or dense, of shape <code class="docutils literal notranslate"><span class="pre">(n_samples,</span> <span class="pre">n_features)</span></code> holding the
training samples, and an array Y of integer values, shape <code class="docutils literal notranslate"><span class="pre">(n_samples,)</span></code>,
holding the class labels for the training samples:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn</span> <span class="kn">import</span> <span class="n">tree</span>
<span class="gp">>>> </span><span class="n">X</span> <span class="o">=</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">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">]]</span>
<span class="gp">>>> </span><span class="n">Y</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span>
<span class="gp">>>> </span><span class="n">clf</span> <span class="o">=</span> <span class="n">tree</span><span class="o">.</span><span class="n">DecisionTreeClassifier</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">clf</span> <span class="o">=</span> <span class="n">clf</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">Y</span><span class="p">)</span>
</pre></div>
</div>
<p>After being fitted, the model can then be used to predict the class of samples:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">clf</span><span class="o">.</span><span class="n">predict</span><span class="p">([[</span><span class="mf">2.</span><span class="p">,</span> <span class="mf">2.</span><span class="p">]])</span>
<span class="go">array([1])</span>
</pre></div>
</div>
<p>In case that there are multiple classes with the same and highest
probability, the classifier will predict the class with the lowest index
amongst those classes.</p>
<p>As an alternative to outputting a specific class, the probability of each class
can be predicted, which is the fraction of training samples of the class in a
leaf:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">clf</span><span class="o">.</span><span class="n">predict_proba</span><span class="p">([[</span><span class="mf">2.</span><span class="p">,</span> <span class="mf">2.</span><span class="p">]])</span>
<span class="go">array([[0., 1.]])</span>
</pre></div>
</div>
<p><a class="reference internal" href="generated/sklearn.tree.DecisionTreeClassifier.html#sklearn.tree.DecisionTreeClassifier" title="sklearn.tree.DecisionTreeClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">DecisionTreeClassifier</span></code></a> is capable of both binary (where the
labels are [-1, 1]) classification and multiclass (where the labels are
[0, …, K-1]) classification.</p>
<p>Using the Iris dataset, we can construct a tree as follows:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.datasets</span> <span class="kn">import</span> <span class="n">load_iris</span>
<span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn</span> <span class="kn">import</span> <span class="n">tree</span>
<span class="gp">>>> </span><span class="n">iris</span> <span class="o">=</span> <span class="n">load_iris</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">iris</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">iris</span><span class="o">.</span><span class="n">target</span>
<span class="gp">>>> </span><span class="n">clf</span> <span class="o">=</span> <span class="n">tree</span><span class="o">.</span><span class="n">DecisionTreeClassifier</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">clf</span> <span class="o">=</span> <span class="n">clf</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
</pre></div>
</div>
<p>Once trained, you can plot the tree with the <a class="reference internal" href="generated/sklearn.tree.plot_tree.html#sklearn.tree.plot_tree" title="sklearn.tree.plot_tree"><code class="xref py py-func docutils literal notranslate"><span class="pre">plot_tree</span></code></a> function:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">tree</span><span class="o">.</span><span class="n">plot_tree</span><span class="p">(</span><span class="n">clf</span><span class="p">)</span>
<span class="go">[...]</span>
</pre></div>
</div>
<figure class="align-center">
<a class="reference external image-reference" href="../auto_examples/tree/plot_iris_dtc.html"><img alt="../_images/sphx_glr_plot_iris_dtc_002.png" src="../_images/sphx_glr_plot_iris_dtc_002.png" style="width: 480.0px; height: 360.0px;" /></a>
</figure>
<p><details id="summary-anchor">
<summary class="btn btn-light">
<strong>Alternative ways to export trees</strong>
<span class="tooltiptext">Click for more details</span>
<a class="headerlink" href="#summary-anchor" title="Permalink to this heading">¶</a>
</summary>
<div class="card"></p>
<p>We can also export the tree in <a class="reference external" href="https://www.graphviz.org/">Graphviz</a> format using the <a class="reference internal" href="generated/sklearn.tree.export_graphviz.html#sklearn.tree.export_graphviz" title="sklearn.tree.export_graphviz"><code class="xref py py-func docutils literal notranslate"><span class="pre">export_graphviz</span></code></a>
exporter. If you use the <a class="reference external" href="https://conda.io">conda</a> package manager, the graphviz binaries
and the python package can be installed with <code class="docutils literal notranslate"><span class="pre">conda</span> <span class="pre">install</span> <span class="pre">python-graphviz</span></code>.</p>
<p>Alternatively binaries for graphviz can be downloaded from the graphviz project homepage,
and the Python wrapper installed from pypi with <code class="docutils literal notranslate"><span class="pre">pip</span> <span class="pre">install</span> <span class="pre">graphviz</span></code>.</p>
<p>Below is an example graphviz export of the above tree trained on the entire
iris dataset; the results are saved in an output file <code class="docutils literal notranslate"><span class="pre">iris.pdf</span></code>:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">graphviz</span>
<span class="gp">>>> </span><span class="n">dot_data</span> <span class="o">=</span> <span class="n">tree</span><span class="o">.</span><span class="n">export_graphviz</span><span class="p">(</span><span class="n">clf</span><span class="p">,</span> <span class="n">out_file</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">graph</span> <span class="o">=</span> <span class="n">graphviz</span><span class="o">.</span><span class="n">Source</span><span class="p">(</span><span class="n">dot_data</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">graph</span><span class="o">.</span><span class="n">render</span><span class="p">(</span><span class="s2">"iris"</span><span class="p">)</span>
</pre></div>
</div>
<p>The <a class="reference internal" href="generated/sklearn.tree.export_graphviz.html#sklearn.tree.export_graphviz" title="sklearn.tree.export_graphviz"><code class="xref py py-func docutils literal notranslate"><span class="pre">export_graphviz</span></code></a> exporter also supports a variety of aesthetic
options, including coloring nodes by their class (or value for regression) and
using explicit variable and class names if desired. Jupyter notebooks also
render these plots inline automatically:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">dot_data</span> <span class="o">=</span> <span class="n">tree</span><span class="o">.</span><span class="n">export_graphviz</span><span class="p">(</span><span class="n">clf</span><span class="p">,</span> <span class="n">out_file</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">feature_names</span><span class="o">=</span><span class="n">iris</span><span class="o">.</span><span class="n">feature_names</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">class_names</span><span class="o">=</span><span class="n">iris</span><span class="o">.</span><span class="n">target_names</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">filled</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">rounded</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">special_characters</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">graph</span> <span class="o">=</span> <span class="n">graphviz</span><span class="o">.</span><span class="n">Source</span><span class="p">(</span><span class="n">dot_data</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">graph</span>
</pre></div>
</div>
<figure class="align-center">
<img alt="../_images/iris.svg" src="../_images/iris.svg" /></figure>
<figure class="align-center">
<a class="reference external image-reference" href="../auto_examples/tree/plot_iris_dtc.html"><img alt="../_images/sphx_glr_plot_iris_dtc_001.png" src="../_images/sphx_glr_plot_iris_dtc_001.png" style="width: 480.0px; height: 360.0px;" /></a>
</figure>
<p>Alternatively, the tree can also be exported in textual format with the
function <a class="reference internal" href="generated/sklearn.tree.export_text.html#sklearn.tree.export_text" title="sklearn.tree.export_text"><code class="xref py py-func docutils literal notranslate"><span class="pre">export_text</span></code></a>. This method doesn’t require the installation
of external libraries and is more compact:</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.datasets</span> <span class="kn">import</span> <span class="n">load_iris</span>
<span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.tree</span> <span class="kn">import</span> <span class="n">DecisionTreeClassifier</span>
<span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.tree</span> <span class="kn">import</span> <span class="n">export_text</span>
<span class="gp">>>> </span><span class="n">iris</span> <span class="o">=</span> <span class="n">load_iris</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">decision_tree</span> <span class="o">=</span> <span class="n">DecisionTreeClassifier</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">max_depth</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">decision_tree</span> <span class="o">=</span> <span class="n">decision_tree</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">iris</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">iris</span><span class="o">.</span><span class="n">target</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">r</span> <span class="o">=</span> <span class="n">export_text</span><span class="p">(</span><span class="n">decision_tree</span><span class="p">,</span> <span class="n">feature_names</span><span class="o">=</span><span class="n">iris</span><span class="p">[</span><span class="s1">'feature_names'</span><span class="p">])</span>
<span class="gp">>>> </span><span class="nb">print</span><span class="p">(</span><span class="n">r</span><span class="p">)</span>
<span class="go">|--- petal width (cm) <= 0.80</span>
<span class="go">| |--- class: 0</span>
<span class="go">|--- petal width (cm) > 0.80</span>
<span class="go">| |--- petal width (cm) <= 1.75</span>
<span class="go">| | |--- class: 1</span>
<span class="go">| |--- petal width (cm) > 1.75</span>
<span class="go">| | |--- class: 2</span>
</pre></div>
</div>
<p></div>
</details></p>
<aside class="topic">
<p class="topic-title">Examples:</p>
<ul class="simple">
<li><p><a class="reference internal" href="../auto_examples/tree/plot_iris_dtc.html#sphx-glr-auto-examples-tree-plot-iris-dtc-py"><span class="std std-ref">Plot the decision surface of decision trees trained on the iris dataset</span></a></p></li>
<li><p><a class="reference internal" href="../auto_examples/tree/plot_unveil_tree_structure.html#sphx-glr-auto-examples-tree-plot-unveil-tree-structure-py"><span class="std std-ref">Understanding the decision tree structure</span></a></p></li>
</ul>
</aside>
</section>
<section id="regression">
<span id="tree-regression"></span><h2><span class="section-number">1.10.2. </span>Regression<a class="headerlink" href="#regression" title="Link to this heading">¶</a></h2>
<figure class="align-center">
<a class="reference external image-reference" href="../auto_examples/tree/plot_tree_regression.html"><img alt="../_images/sphx_glr_plot_tree_regression_001.png" src="../_images/sphx_glr_plot_tree_regression_001.png" style="width: 480.0px; height: 360.0px;" /></a>
</figure>
<p>Decision trees can also be applied to regression problems, using the
<a class="reference internal" href="generated/sklearn.tree.DecisionTreeRegressor.html#sklearn.tree.DecisionTreeRegressor" title="sklearn.tree.DecisionTreeRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">DecisionTreeRegressor</span></code></a> class.</p>
<p>As in the classification setting, the fit method will take as argument arrays X
and y, only that in this case y is expected to have floating point values
instead of integer values:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn</span> <span class="kn">import</span> <span class="n">tree</span>
<span class="gp">>>> </span><span class="n">X</span> <span class="o">=</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">2</span><span class="p">,</span> <span class="mi">2</span><span class="p">]]</span>
<span class="gp">>>> </span><span class="n">y</span> <span class="o">=</span> <span class="p">[</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">2.5</span><span class="p">]</span>
<span class="gp">>>> </span><span class="n">clf</span> <span class="o">=</span> <span class="n">tree</span><span class="o">.</span><span class="n">DecisionTreeRegressor</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">clf</span> <span class="o">=</span> <span class="n">clf</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">clf</span><span class="o">.</span><span class="n">predict</span><span class="p">([[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">]])</span>
<span class="go">array([0.5])</span>
</pre></div>
</div>
<aside class="topic">
<p class="topic-title">Examples:</p>
<ul class="simple">
<li><p><a class="reference internal" href="../auto_examples/tree/plot_tree_regression.html#sphx-glr-auto-examples-tree-plot-tree-regression-py"><span class="std std-ref">Decision Tree Regression</span></a></p></li>
</ul>
</aside>
</section>
<section id="multi-output-problems">
<span id="tree-multioutput"></span><h2><span class="section-number">1.10.3. </span>Multi-output problems<a class="headerlink" href="#multi-output-problems" title="Link to this heading">¶</a></h2>
<p>A multi-output problem is a supervised learning problem with several outputs
to predict, that is when Y is a 2d array of shape <code class="docutils literal notranslate"><span class="pre">(n_samples,</span> <span class="pre">n_outputs)</span></code>.</p>
<p>When there is no correlation between the outputs, a very simple way to solve
this kind of problem is to build n independent models, i.e. one for each
output, and then to use those models to independently predict each one of the n
outputs. However, because it is likely that the output values related to the
same input are themselves correlated, an often better way is to build a single
model capable of predicting simultaneously all n outputs. First, it requires
lower training time since only a single estimator is built. Second, the
generalization accuracy of the resulting estimator may often be increased.</p>
<p>With regard to decision trees, this strategy can readily be used to support
multi-output problems. This requires the following changes:</p>
<ul class="simple">
<li><p>Store n output values in leaves, instead of 1;</p></li>
<li><p>Use splitting criteria that compute the average reduction across all
n outputs.</p></li>
</ul>
<p>This module offers support for multi-output problems by implementing this
strategy in both <a class="reference internal" href="generated/sklearn.tree.DecisionTreeClassifier.html#sklearn.tree.DecisionTreeClassifier" title="sklearn.tree.DecisionTreeClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">DecisionTreeClassifier</span></code></a> and
<a class="reference internal" href="generated/sklearn.tree.DecisionTreeRegressor.html#sklearn.tree.DecisionTreeRegressor" title="sklearn.tree.DecisionTreeRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">DecisionTreeRegressor</span></code></a>. If a decision tree is fit on an output array Y
of shape <code class="docutils literal notranslate"><span class="pre">(n_samples,</span> <span class="pre">n_outputs)</span></code> then the resulting estimator will:</p>
<ul class="simple">
<li><p>Output n_output values upon <code class="docutils literal notranslate"><span class="pre">predict</span></code>;</p></li>
<li><p>Output a list of n_output arrays of class probabilities upon
<code class="docutils literal notranslate"><span class="pre">predict_proba</span></code>.</p></li>
</ul>
<p>The use of multi-output trees for regression is demonstrated in
<a class="reference internal" href="../auto_examples/tree/plot_tree_regression_multioutput.html#sphx-glr-auto-examples-tree-plot-tree-regression-multioutput-py"><span class="std std-ref">Multi-output Decision Tree Regression</span></a>. In this example, the input
X is a single real value and the outputs Y are the sine and cosine of X.</p>
<figure class="align-center">
<a class="reference external image-reference" href="../auto_examples/tree/plot_tree_regression_multioutput.html"><img alt="../_images/sphx_glr_plot_tree_regression_multioutput_001.png" src="../_images/sphx_glr_plot_tree_regression_multioutput_001.png" style="width: 480.0px; height: 360.0px;" /></a>
</figure>
<p>The use of multi-output trees for classification is demonstrated in
<a class="reference internal" href="../auto_examples/miscellaneous/plot_multioutput_face_completion.html#sphx-glr-auto-examples-miscellaneous-plot-multioutput-face-completion-py"><span class="std std-ref">Face completion with a multi-output estimators</span></a>. In this example, the inputs
X are the pixels of the upper half of faces and the outputs Y are the pixels of
the lower half of those faces.</p>
<figure class="align-center">
<a class="reference external image-reference" href="../auto_examples/miscellaneous/plot_multioutput_face_completion.html"><img alt="../_images/sphx_glr_plot_multioutput_face_completion_001.png" src="../_images/sphx_glr_plot_multioutput_face_completion_001.png" style="width: 750.0px; height: 847.5px;" /></a>
</figure>
<aside class="topic">
<p class="topic-title">Examples:</p>
<ul class="simple">
<li><p><a class="reference internal" href="../auto_examples/tree/plot_tree_regression_multioutput.html#sphx-glr-auto-examples-tree-plot-tree-regression-multioutput-py"><span class="std std-ref">Multi-output Decision Tree Regression</span></a></p></li>
<li><p><a class="reference internal" href="../auto_examples/miscellaneous/plot_multioutput_face_completion.html#sphx-glr-auto-examples-miscellaneous-plot-multioutput-face-completion-py"><span class="std std-ref">Face completion with a multi-output estimators</span></a></p></li>
</ul>
</aside>
<p><details id="summary-anchor">
<summary class="btn btn-light">
<strong>References</strong>
<span class="tooltiptext">Click for more details</span>
<a class="headerlink" href="#summary-anchor" title="Permalink to this heading">¶</a>
</summary>
<div class="card"></p>
<ul class="simple">
<li><p>M. Dumont et al, <a class="reference external" href="http://www.montefiore.ulg.ac.be/services/stochastic/pubs/2009/DMWG09/dumont-visapp09-shortpaper.pdf">Fast multi-class image annotation with random subwindows
and multiple output randomized trees</a>, International Conference on
Computer Vision Theory and Applications 2009</p></li>
</ul>
<p></div>
</details></p>
</section>
<section id="complexity">
<span id="tree-complexity"></span><h2><span class="section-number">1.10.4. </span>Complexity<a class="headerlink" href="#complexity" title="Link to this heading">¶</a></h2>
<p>In general, the run time cost to construct a balanced binary tree is
<span class="math notranslate nohighlight">\(O(n_{samples}n_{features}\log(n_{samples}))\)</span> and query time
<span class="math notranslate nohighlight">\(O(\log(n_{samples}))\)</span>. Although the tree construction algorithm attempts
to generate balanced trees, they will not always be balanced. Assuming that the
subtrees remain approximately balanced, the cost at each node consists of
searching through <span class="math notranslate nohighlight">\(O(n_{features})\)</span> to find the feature that offers the
largest reduction in the impurity criterion, e.g. log loss (which is equivalent to an
information gain). This has a cost of
<span class="math notranslate nohighlight">\(O(n_{features}n_{samples}\log(n_{samples}))\)</span> at each node, leading to a
total cost over the entire trees (by summing the cost at each node) of
<span class="math notranslate nohighlight">\(O(n_{features}n_{samples}^{2}\log(n_{samples}))\)</span>.</p>
</section>
<section id="tips-on-practical-use">
<h2><span class="section-number">1.10.5. </span>Tips on practical use<a class="headerlink" href="#tips-on-practical-use" title="Link to this heading">¶</a></h2>
<ul>
<li><p>Decision trees tend to overfit on data with a large number of features.
Getting the right ratio of samples to number of features is important, since
a tree with few samples in high dimensional space is very likely to overfit.</p></li>
<li><p>Consider performing dimensionality reduction (<a class="reference internal" href="decomposition.html#pca"><span class="std std-ref">PCA</span></a>,
<a class="reference internal" href="decomposition.html#ica"><span class="std std-ref">ICA</span></a>, or <a class="reference internal" href="feature_selection.html#feature-selection"><span class="std std-ref">Feature selection</span></a>) beforehand to
give your tree a better chance of finding features that are discriminative.</p></li>
<li><p><a class="reference internal" href="../auto_examples/tree/plot_unveil_tree_structure.html#sphx-glr-auto-examples-tree-plot-unveil-tree-structure-py"><span class="std std-ref">Understanding the decision tree structure</span></a> will help
in gaining more insights about how the decision tree makes predictions, which is
important for understanding the important features in the data.</p></li>
<li><p>Visualize your tree as you are training by using the <code class="docutils literal notranslate"><span class="pre">export</span></code>
function. Use <code class="docutils literal notranslate"><span class="pre">max_depth=3</span></code> as an initial tree depth to get a feel for
how the tree is fitting to your data, and then increase the depth.</p></li>
<li><p>Remember that the number of samples required to populate the tree doubles
for each additional level the tree grows to. Use <code class="docutils literal notranslate"><span class="pre">max_depth</span></code> to control
the size of the tree to prevent overfitting.</p></li>
<li><p>Use <code class="docutils literal notranslate"><span class="pre">min_samples_split</span></code> or <code class="docutils literal notranslate"><span class="pre">min_samples_leaf</span></code> to ensure that multiple
samples inform every decision in the tree, by controlling which splits will
be considered. A very small number will usually mean the tree will overfit,
whereas a large number will prevent the tree from learning the data. Try
<code class="docutils literal notranslate"><span class="pre">min_samples_leaf=5</span></code> as an initial value. If the sample size varies
greatly, a float number can be used as percentage in these two parameters.
While <code class="docutils literal notranslate"><span class="pre">min_samples_split</span></code> can create arbitrarily small leaves,
<code class="docutils literal notranslate"><span class="pre">min_samples_leaf</span></code> guarantees that each leaf has a minimum size, avoiding
low-variance, over-fit leaf nodes in regression problems. For
classification with few classes, <code class="docutils literal notranslate"><span class="pre">min_samples_leaf=1</span></code> is often the best
choice.</p>
<p>Note that <code class="docutils literal notranslate"><span class="pre">min_samples_split</span></code> considers samples directly and independent of
<code class="docutils literal notranslate"><span class="pre">sample_weight</span></code>, if provided (e.g. a node with m weighted samples is still
treated as having exactly m samples). Consider <code class="docutils literal notranslate"><span class="pre">min_weight_fraction_leaf</span></code> or
<code class="docutils literal notranslate"><span class="pre">min_impurity_decrease</span></code> if accounting for sample weights is required at splits.</p>
</li>
<li><p>Balance your dataset before training to prevent the tree from being biased
toward the classes that are dominant. Class balancing can be done by
sampling an equal number of samples from each class, or preferably by
normalizing the sum of the sample weights (<code class="docutils literal notranslate"><span class="pre">sample_weight</span></code>) for each
class to the same value. Also note that weight-based pre-pruning criteria,
such as <code class="docutils literal notranslate"><span class="pre">min_weight_fraction_leaf</span></code>, will then be less biased toward
dominant classes than criteria that are not aware of the sample weights,
like <code class="docutils literal notranslate"><span class="pre">min_samples_leaf</span></code>.</p></li>
<li><p>If the samples are weighted, it will be easier to optimize the tree
structure using weight-based pre-pruning criterion such as
<code class="docutils literal notranslate"><span class="pre">min_weight_fraction_leaf</span></code>, which ensure that leaf nodes contain at least
a fraction of the overall sum of the sample weights.</p></li>
<li><p>All decision trees use <code class="docutils literal notranslate"><span class="pre">np.float32</span></code> arrays internally.
If training data is not in this format, a copy of the dataset will be made.</p></li>
<li><p>If the input matrix X is very sparse, it is recommended to convert to sparse
<code class="docutils literal notranslate"><span class="pre">csc_matrix</span></code> before calling fit and sparse <code class="docutils literal notranslate"><span class="pre">csr_matrix</span></code> before calling
predict. Training time can be orders of magnitude faster for a sparse
matrix input compared to a dense matrix when features have zero values in
most of the samples.</p></li>
</ul>
</section>
<section id="tree-algorithms-id3-c4-5-c5-0-and-cart">
<span id="tree-algorithms"></span><h2><span class="section-number">1.10.6. </span>Tree algorithms: ID3, C4.5, C5.0 and CART<a class="headerlink" href="#tree-algorithms-id3-c4-5-c5-0-and-cart" title="Link to this heading">¶</a></h2>
<p>What are all the various decision tree algorithms and how do they differ
from each other? Which one is implemented in scikit-learn?</p>
<p><details id="summary-anchor">
<summary class="btn btn-light">
<strong>Various decision tree algorithms</strong>
<span class="tooltiptext">Click for more details</span>
<a class="headerlink" href="#summary-anchor" title="Permalink to this heading">¶</a>
</summary>
<div class="card"></p>
<p><a class="reference external" href="https://en.wikipedia.org/wiki/ID3_algorithm">ID3</a> (Iterative Dichotomiser 3) was developed in 1986 by Ross Quinlan.
The algorithm creates a multiway tree, finding for each node (i.e. in
a greedy manner) the categorical feature that will yield the largest
information gain for categorical targets. Trees are grown to their
maximum size and then a pruning step is usually applied to improve the
ability of the tree to generalize to unseen data.</p>
<p>C4.5 is the successor to ID3 and removed the restriction that features
must be categorical by dynamically defining a discrete attribute (based
on numerical variables) that partitions the continuous attribute value
into a discrete set of intervals. C4.5 converts the trained trees
(i.e. the output of the ID3 algorithm) into sets of if-then rules.
The accuracy of each rule is then evaluated to determine the order
in which they should be applied. Pruning is done by removing a rule’s
precondition if the accuracy of the rule improves without it.</p>
<p>C5.0 is Quinlan’s latest version release under a proprietary license.
It uses less memory and builds smaller rulesets than C4.5 while being
more accurate.</p>
<p>CART (Classification and Regression Trees) is very similar to C4.5, but
it differs in that it supports numerical target variables (regression) and
does not compute rule sets. CART constructs binary trees using the feature
and threshold that yield the largest information gain at each node.</p>
<p></div>
</details></p>
<p>scikit-learn uses an optimized version of the CART algorithm; however, the
scikit-learn implementation does not support categorical variables for now.</p>
</section>
<section id="mathematical-formulation">
<span id="tree-mathematical-formulation"></span><h2><span class="section-number">1.10.7. </span>Mathematical formulation<a class="headerlink" href="#mathematical-formulation" title="Link to this heading">¶</a></h2>
<p>Given training vectors <span class="math notranslate nohighlight">\(x_i \in R^n\)</span>, i=1,…, l and a label vector
<span class="math notranslate nohighlight">\(y \in R^l\)</span>, a decision tree recursively partitions the feature space
such that the samples with the same labels or similar target values are grouped
together.</p>
<p>Let the data at node <span class="math notranslate nohighlight">\(m\)</span> be represented by <span class="math notranslate nohighlight">\(Q_m\)</span> with <span class="math notranslate nohighlight">\(n_m\)</span>
samples. For each candidate split <span class="math notranslate nohighlight">\(\theta = (j, t_m)\)</span> consisting of a
feature <span class="math notranslate nohighlight">\(j\)</span> and threshold <span class="math notranslate nohighlight">\(t_m\)</span>, partition the data into
<span class="math notranslate nohighlight">\(Q_m^{left}(\theta)\)</span> and <span class="math notranslate nohighlight">\(Q_m^{right}(\theta)\)</span> subsets</p>
<div class="math notranslate nohighlight">
\[ \begin{align}\begin{aligned}Q_m^{left}(\theta) = \{(x, y) | x_j \leq t_m\}\\Q_m^{right}(\theta) = Q_m \setminus Q_m^{left}(\theta)\end{aligned}\end{align} \]</div>
<p>The quality of a candidate split of node <span class="math notranslate nohighlight">\(m\)</span> is then computed using an
impurity function or loss function <span class="math notranslate nohighlight">\(H()\)</span>, the choice of which depends on
the task being solved (classification or regression)</p>
<div class="math notranslate nohighlight">
\[G(Q_m, \theta) = \frac{n_m^{left}}{n_m} H(Q_m^{left}(\theta))
+ \frac{n_m^{right}}{n_m} H(Q_m^{right}(\theta))\]</div>
<p>Select the parameters that minimises the impurity</p>
<div class="math notranslate nohighlight">
\[\theta^* = \operatorname{argmin}_\theta G(Q_m, \theta)\]</div>
<p>Recurse for subsets <span class="math notranslate nohighlight">\(Q_m^{left}(\theta^*)\)</span> and
<span class="math notranslate nohighlight">\(Q_m^{right}(\theta^*)\)</span> until the maximum allowable depth is reached,
<span class="math notranslate nohighlight">\(n_m < \min_{samples}\)</span> or <span class="math notranslate nohighlight">\(n_m = 1\)</span>.</p>
<section id="classification-criteria">
<h3><span class="section-number">1.10.7.1. </span>Classification criteria<a class="headerlink" href="#classification-criteria" title="Link to this heading">¶</a></h3>
<p>If a target is a classification outcome taking on values 0,1,…,K-1,
for node <span class="math notranslate nohighlight">\(m\)</span>, let</p>
<div class="math notranslate nohighlight">
\[p_{mk} = \frac{1}{n_m} \sum_{y \in Q_m} I(y = k)\]</div>
<p>be the proportion of class k observations in node <span class="math notranslate nohighlight">\(m\)</span>. If <span class="math notranslate nohighlight">\(m\)</span> is a
terminal node, <code class="docutils literal notranslate"><span class="pre">predict_proba</span></code> for this region is set to <span class="math notranslate nohighlight">\(p_{mk}\)</span>.
Common measures of impurity are the following.</p>
<p>Gini:</p>
<div class="math notranslate nohighlight">
\[H(Q_m) = \sum_k p_{mk} (1 - p_{mk})\]</div>
<p>Log Loss or Entropy:</p>
<div class="math notranslate nohighlight">
\[H(Q_m) = - \sum_k p_{mk} \log(p_{mk})\]</div>
<p><details id="summary-anchor">
<summary class="btn btn-light">
<strong>Shannon entropy</strong>
<span class="tooltiptext">Click for more details</span>
<a class="headerlink" href="#summary-anchor" title="Permalink to this heading">¶</a>
</summary>
<div class="card"></p>
<p>The entropy criterion computes the Shannon entropy of the possible classes. It
takes the class frequencies of the training data points that reached a given
leaf <span class="math notranslate nohighlight">\(m\)</span> as their probability. Using the <strong>Shannon entropy as tree node
splitting criterion is equivalent to minimizing the log loss</strong> (also known as
cross-entropy and multinomial deviance) between the true labels <span class="math notranslate nohighlight">\(y_i\)</span>
and the probabilistic predictions <span class="math notranslate nohighlight">\(T_k(x_i)\)</span> of the tree model <span class="math notranslate nohighlight">\(T\)</span> for class <span class="math notranslate nohighlight">\(k\)</span>.</p>
<p>To see this, first recall that the log loss of a tree model <span class="math notranslate nohighlight">\(T\)</span>
computed on a dataset <span class="math notranslate nohighlight">\(D\)</span> is defined as follows:</p>
<div class="math notranslate nohighlight">
\[\mathrm{LL}(D, T) = -\frac{1}{n} \sum_{(x_i, y_i) \in D} \sum_k I(y_i = k) \log(T_k(x_i))\]</div>
<p>where <span class="math notranslate nohighlight">\(D\)</span> is a training dataset of <span class="math notranslate nohighlight">\(n\)</span> pairs <span class="math notranslate nohighlight">\((x_i, y_i)\)</span>.</p>
<p>In a classification tree, the predicted class probabilities within leaf nodes
are constant, that is: for all <span class="math notranslate nohighlight">\((x_i, y_i) \in Q_m\)</span>, one has:
<span class="math notranslate nohighlight">\(T_k(x_i) = p_{mk}\)</span> for each class <span class="math notranslate nohighlight">\(k\)</span>.</p>
<p>This property makes it possible to rewrite <span class="math notranslate nohighlight">\(\mathrm{LL}(D, T)\)</span> as the
sum of the Shannon entropies computed for each leaf of <span class="math notranslate nohighlight">\(T\)</span> weighted by
the number of training data points that reached each leaf:</p>
<div class="math notranslate nohighlight">
\[\mathrm{LL}(D, T) = \sum_{m \in T} \frac{n_m}{n} H(Q_m)\]</div>
<p></div>
</details></p>
</section>
<section id="regression-criteria">
<h3><span class="section-number">1.10.7.2. </span>Regression criteria<a class="headerlink" href="#regression-criteria" title="Link to this heading">¶</a></h3>
<p>If the target is a continuous value, then for node <span class="math notranslate nohighlight">\(m\)</span>, common
criteria to minimize as for determining locations for future splits are Mean
Squared Error (MSE or L2 error), Poisson deviance as well as Mean Absolute
Error (MAE or L1 error). MSE and Poisson deviance both set the predicted value
of terminal nodes to the learned mean value <span class="math notranslate nohighlight">\(\bar{y}_m\)</span> of the node
whereas the MAE sets the predicted value of terminal nodes to the median
<span class="math notranslate nohighlight">\(median(y)_m\)</span>.</p>
<p>Mean Squared Error:</p>
<div class="math notranslate nohighlight">
\[ \begin{align}\begin{aligned}\bar{y}_m = \frac{1}{n_m} \sum_{y \in Q_m} y\\H(Q_m) = \frac{1}{n_m} \sum_{y \in Q_m} (y - \bar{y}_m)^2\end{aligned}\end{align} \]</div>
<p>Half Poisson deviance:</p>
<div class="math notranslate nohighlight">
\[H(Q_m) = \frac{1}{n_m} \sum_{y \in Q_m} (y \log\frac{y}{\bar{y}_m}
- y + \bar{y}_m)\]</div>
<p>Setting <code class="docutils literal notranslate"><span class="pre">criterion="poisson"</span></code> might be a good choice if your target is a count
or a frequency (count per some unit). In any case, <span class="math notranslate nohighlight">\(y >= 0\)</span> is a
necessary condition to use this criterion. Note that it fits much slower than
the MSE criterion.</p>
<p>Mean Absolute Error:</p>
<div class="math notranslate nohighlight">
\[ \begin{align}\begin{aligned}median(y)_m = \underset{y \in Q_m}{\mathrm{median}}(y)\\H(Q_m) = \frac{1}{n_m} \sum_{y \in Q_m} |y - median(y)_m|\end{aligned}\end{align} \]</div>
<p>Note that it fits much slower than the MSE criterion.</p>
</section>
</section>
<section id="missing-values-support">
<span id="tree-missing-value-support"></span><h2><span class="section-number">1.10.8. </span>Missing Values Support<a class="headerlink" href="#missing-values-support" title="Link to this heading">¶</a></h2>
<p><a class="reference internal" href="generated/sklearn.tree.DecisionTreeClassifier.html#sklearn.tree.DecisionTreeClassifier" title="sklearn.tree.DecisionTreeClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">DecisionTreeClassifier</span></code></a> and <a class="reference internal" href="generated/sklearn.tree.DecisionTreeRegressor.html#sklearn.tree.DecisionTreeRegressor" title="sklearn.tree.DecisionTreeRegressor"><code class="xref py py-class docutils literal notranslate"><span class="pre">DecisionTreeRegressor</span></code></a>
have built-in support for missing values when <code class="docutils literal notranslate"><span class="pre">splitter='best'</span></code> and criterion is
<code class="docutils literal notranslate"><span class="pre">'gini'</span></code>, <code class="docutils literal notranslate"><span class="pre">'entropy</span></code>’, or <code class="docutils literal notranslate"><span class="pre">'log_loss'</span></code>, for classification or
<code class="docutils literal notranslate"><span class="pre">'squared_error'</span></code>, <code class="docutils literal notranslate"><span class="pre">'friedman_mse'</span></code>, or <code class="docutils literal notranslate"><span class="pre">'poisson'</span></code> for regression.</p>
<p>For each potential threshold on the non-missing data, the splitter will evaluate
the split with all the missing values going to the left node or the right node.</p>
<p>Decisions are made as follows:</p>
<ul>
<li><p>By default when predicting, the samples with missing values are classified
with the class used in the split found during training:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.tree</span> <span class="kn">import</span> <span class="n">DecisionTreeClassifier</span>
<span class="gp">>>> </span><span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
<span class="gp">>>> </span><span class="n">X</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">6</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">nan</span><span class="p">])</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">y</span> <span class="o">=</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="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span>
<span class="gp">>>> </span><span class="n">tree</span> <span class="o">=</span> <span class="n">DecisionTreeClassifier</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="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">tree</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X</span><span class="p">)</span>
<span class="go">array([0, 0, 1, 1])</span>
</pre></div>
</div>
</li>
<li><p>If the criterion evaluation is the same for both nodes,
then the tie for missing value at predict time is broken by going to the
right node. The splitter also checks the split where all the missing
values go to one child and non-missing values go to the other:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.tree</span> <span class="kn">import</span> <span class="n">DecisionTreeClassifier</span>
<span class="gp">>>> </span><span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
<span class="gp">>>> </span><span class="n">X</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([</span><span class="n">np</span><span class="o">.</span><span class="n">nan</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">nan</span><span class="p">,</span> <span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">y</span> <span class="o">=</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="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span>
<span class="gp">>>> </span><span class="n">tree</span> <span class="o">=</span> <span class="n">DecisionTreeClassifier</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="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_test</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([</span><span class="n">np</span><span class="o">.</span><span class="n">nan</span><span class="p">])</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">tree</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
<span class="go">array([1])</span>
</pre></div>
</div>
</li>
<li><p>If no missing values are seen during training for a given feature, then during
prediction missing values are mapped to the child with the most samples:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.tree</span> <span class="kn">import</span> <span class="n">DecisionTreeClassifier</span>
<span class="gp">>>> </span><span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
<span class="gp">>>> </span><span class="n">X</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([</span><span class="mi">0</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="mi">3</span><span class="p">])</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">y</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span>
<span class="gp">>>> </span><span class="n">tree</span> <span class="o">=</span> <span class="n">DecisionTreeClassifier</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="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_test</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([</span><span class="n">np</span><span class="o">.</span><span class="n">nan</span><span class="p">])</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">tree</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
<span class="go">array([1])</span>
</pre></div>
</div>
</li>
</ul>
</section>
<section id="minimal-cost-complexity-pruning">
<span id="id1"></span><h2><span class="section-number">1.10.9. </span>Minimal Cost-Complexity Pruning<a class="headerlink" href="#minimal-cost-complexity-pruning" title="Link to this heading">¶</a></h2>
<p>Minimal cost-complexity pruning is an algorithm used to prune a tree to avoid
over-fitting, described in Chapter 3 of <a class="reference internal" href="#bre" id="id2"><span>[BRE]</span></a>. This algorithm is parameterized
by <span class="math notranslate nohighlight">\(\alpha\ge0\)</span> known as the complexity parameter. The complexity
parameter is used to define the cost-complexity measure, <span class="math notranslate nohighlight">\(R_\alpha(T)\)</span> of
a given tree <span class="math notranslate nohighlight">\(T\)</span>:</p>
<div class="math notranslate nohighlight">
\[R_\alpha(T) = R(T) + \alpha|\widetilde{T}|\]</div>
<p>where <span class="math notranslate nohighlight">\(|\widetilde{T}|\)</span> is the number of terminal nodes in <span class="math notranslate nohighlight">\(T\)</span> and <span class="math notranslate nohighlight">\(R(T)\)</span>
is traditionally defined as the total misclassification rate of the terminal
nodes. Alternatively, scikit-learn uses the total sample weighted impurity of
the terminal nodes for <span class="math notranslate nohighlight">\(R(T)\)</span>. As shown above, the impurity of a node
depends on the criterion. Minimal cost-complexity pruning finds the subtree of
<span class="math notranslate nohighlight">\(T\)</span> that minimizes <span class="math notranslate nohighlight">\(R_\alpha(T)\)</span>.</p>
<p>The cost complexity measure of a single node is
<span class="math notranslate nohighlight">\(R_\alpha(t)=R(t)+\alpha\)</span>. The branch, <span class="math notranslate nohighlight">\(T_t\)</span>, is defined to be a
tree where node <span class="math notranslate nohighlight">\(t\)</span> is its root. In general, the impurity of a node
is greater than the sum of impurities of its terminal nodes,
<span class="math notranslate nohighlight">\(R(T_t)<R(t)\)</span>. However, the cost complexity measure of a node,
<span class="math notranslate nohighlight">\(t\)</span>, and its branch, <span class="math notranslate nohighlight">\(T_t\)</span>, can be equal depending on
<span class="math notranslate nohighlight">\(\alpha\)</span>. We define the effective <span class="math notranslate nohighlight">\(\alpha\)</span> of a node to be the
value where they are equal, <span class="math notranslate nohighlight">\(R_\alpha(T_t)=R_\alpha(t)\)</span> or
<span class="math notranslate nohighlight">\(\alpha_{eff}(t)=\frac{R(t)-R(T_t)}{|T|-1}\)</span>. A non-terminal node
with the smallest value of <span class="math notranslate nohighlight">\(\alpha_{eff}\)</span> is the weakest link and will
be pruned. This process stops when the pruned tree’s minimal
<span class="math notranslate nohighlight">\(\alpha_{eff}\)</span> is greater than the <code class="docutils literal notranslate"><span class="pre">ccp_alpha</span></code> parameter.</p>
<aside class="topic">
<p class="topic-title">Examples:</p>
<ul class="simple">
<li><p><a class="reference internal" href="../auto_examples/tree/plot_cost_complexity_pruning.html#sphx-glr-auto-examples-tree-plot-cost-complexity-pruning-py"><span class="std std-ref">Post pruning decision trees with cost complexity pruning</span></a></p></li>
</ul>
</aside>
<p><details id="summary-anchor">
<summary class="btn btn-light">
<strong>References</strong>
<span class="tooltiptext">Click for more details</span>
<a class="headerlink" href="#summary-anchor" title="Permalink to this heading">¶</a>
</summary>
<div class="card"></p>
<div role="list" class="citation-list">
<div class="citation" id="bre" role="doc-biblioentry">
<span class="label"><span class="fn-bracket">[</span><a role="doc-backlink" href="#id2">BRE</a><span class="fn-bracket">]</span></span>
<p>L. Breiman, J. Friedman, R. Olshen, and C. Stone. Classification
and Regression Trees. Wadsworth, Belmont, CA, 1984.</p>
</div>
</div>
<ul class="simple">
<li><p><a class="reference external" href="https://en.wikipedia.org/wiki/Decision_tree_learning">https://en.wikipedia.org/wiki/Decision_tree_learning</a></p></li>
<li><p><a class="reference external" href="https://en.wikipedia.org/wiki/Predictive_analytics">https://en.wikipedia.org/wiki/Predictive_analytics</a></p></li>
<li><p>J.R. Quinlan. C4. 5: programs for machine learning. Morgan
Kaufmann, 1993.</p></li>
<li><p>T. Hastie, R. Tibshirani and J. Friedman. Elements of Statistical
Learning, Springer, 2009.</p></li>
</ul>
<p></div>
</details></p>
</section>
</section>
</div>
<div class="container">
<footer class="sk-content-footer">
© 2007 - 2024, scikit-learn developers (BSD License).
<a href="../_sources/modules/tree.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 src="../_static/clipboard.min.js"></script>
<script src="../_static/copybutton.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');
// create and add the button to all the code blocks that contain >>>
div.each(function(index) {
var jthis = $(this);
// 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>');
});
/*** 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>