-
Notifications
You must be signed in to change notification settings - Fork 47
/
naivebayes.html
399 lines (309 loc) · 37.9 KB
/
naivebayes.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
<!DOCTYPE html>
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<meta name="theme-color" content="#2D2D2D" />
<title>NLTK :: nltk.classify.naivebayes</title>
<link rel="stylesheet" href="../../../_static/css/nltk_theme.css"/>
<link rel="stylesheet" href="../../../_static/css/custom.css"/>
<script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
<script type="text/javascript" src="../../../_static/documentation_options.js"></script>
<script type="text/javascript" src="../../../_static/jquery.js"></script>
<script type="text/javascript" src="../../../_static/underscore.js"></script>
<script type="text/javascript" src="../../../_static/doctools.js"></script>
<script src="https://email.tl.fortawesome.com/c/eJxNjUEOgyAQAF8jR7Kw6wIHDh7sP1Cw2mgxgmn6-3JsMqc5zEQfE8dkxOY1KKMUOI3ACFKRJpSW2AAp7ontYIaxI6i7XPJVwyeVfCQ550Os3jLrGSNOLgbdAy6s0PBk2TFNjEbsfq31LB0OnX407pJa5v2faRadwSW63mn5KuLyR9j2tgx3zecanl-55R_-jjPs"></script>
</head>
<body>
<div id="nltk-theme-container">
<header>
<div id="logo-container">
<h1>
<a href="../../../index.html">NLTK</a>
</h1>
</div>
<div id="project-container">
<h1>Documentation</h1>
</div>
<a id="menu-toggle" class="fa fa-bars" aria-hidden="true"></a>
<script type="text/javascript">
$("#menu-toggle").click(function() {
$("#menu-toggle").toggleClass("toggled");
$("#side-menu-container").slideToggle(300);
});
</script>
</header>
<div id="content-container">
<div id="side-menu-container">
<div id="search" role="search">
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
<input type="text" name="q" placeholder="Search" />
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form>
</div>
<div id="side-menu" role="navigation">
<p class="caption" role="heading"><span class="caption-text">NLTK Documentation</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../../api/nltk.html">API Reference</a></li>
<li class="toctree-l1"><a class="reference external" href="http://www.nltk.org/howto">Example Usage</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../py-modindex.html">Module Index</a></li>
<li class="toctree-l1"><a class="reference external" href="https://github.com/nltk/nltk/wiki">Wiki</a></li>
<li class="toctree-l1"><a class="reference external" href="https://github.com/nltk/nltk/wiki/FAQ">FAQ</a></li>
</ul>
<p class="caption" role="heading"><span class="caption-text">Installation</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../../install.html">Installing NLTK</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../data.html">Installing NLTK Data</a></li>
</ul>
<p class="caption" role="heading"><span class="caption-text">More</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../../news.html">Release Notes</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../contribute.html">Contributing to NLTK</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../team.html">NLTK Team</a></li>
</ul>
</div>
</div>
<div id="main-content-container">
<div id="main-content" role="main">
<h1>Source code for nltk.classify.naivebayes</h1><div class="highlight"><pre>
<span></span><span class="c1"># Natural Language Toolkit: Naive Bayes Classifiers</span>
<span class="c1">#</span>
<span class="c1"># Copyright (C) 2001-2021 NLTK Project</span>
<span class="c1"># Author: Edward Loper <edloper@gmail.com></span>
<span class="c1"># URL: <http://nltk.org/></span>
<span class="c1"># For license information, see LICENSE.TXT</span>
<span class="sd">"""</span>
<span class="sd">A classifier based on the Naive Bayes algorithm. In order to find the</span>
<span class="sd">probability for a label, this algorithm first uses the Bayes rule to</span>
<span class="sd">express P(label|features) in terms of P(label) and P(features|label):</span>
<span class="sd">| P(label) * P(features|label)</span>
<span class="sd">| P(label|features) = ------------------------------</span>
<span class="sd">| P(features)</span>
<span class="sd">The algorithm then makes the 'naive' assumption that all features are</span>
<span class="sd">independent, given the label:</span>
<span class="sd">| P(label) * P(f1|label) * ... * P(fn|label)</span>
<span class="sd">| P(label|features) = --------------------------------------------</span>
<span class="sd">| P(features)</span>
<span class="sd">Rather than computing P(features) explicitly, the algorithm just</span>
<span class="sd">calculates the numerator for each label, and normalizes them so they</span>
<span class="sd">sum to one:</span>
<span class="sd">| P(label) * P(f1|label) * ... * P(fn|label)</span>
<span class="sd">| P(label|features) = --------------------------------------------</span>
<span class="sd">| SUM[l]( P(l) * P(f1|l) * ... * P(fn|l) )</span>
<span class="sd">"""</span>
<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">defaultdict</span>
<span class="kn">from</span> <span class="nn">nltk.classify.api</span> <span class="kn">import</span> <span class="n">ClassifierI</span>
<span class="kn">from</span> <span class="nn">nltk.probability</span> <span class="kn">import</span> <span class="n">DictionaryProbDist</span><span class="p">,</span> <span class="n">ELEProbDist</span><span class="p">,</span> <span class="n">FreqDist</span><span class="p">,</span> <span class="n">sum_logs</span>
<span class="c1">##//////////////////////////////////////////////////////</span>
<span class="c1">## Naive Bayes Classifier</span>
<span class="c1">##//////////////////////////////////////////////////////</span>
<div class="viewcode-block" id="NaiveBayesClassifier"><a class="viewcode-back" href="../../../api/nltk.classify.naivebayes.html#nltk.classify.naivebayes.NaiveBayesClassifier">[docs]</a><span class="k">class</span> <span class="nc">NaiveBayesClassifier</span><span class="p">(</span><span class="n">ClassifierI</span><span class="p">):</span>
<span class="sd">"""</span>
<span class="sd"> A Naive Bayes classifier. Naive Bayes classifiers are</span>
<span class="sd"> paramaterized by two probability distributions:</span>
<span class="sd"> - P(label) gives the probability that an input will receive each</span>
<span class="sd"> label, given no information about the input's features.</span>
<span class="sd"> - P(fname=fval|label) gives the probability that a given feature</span>
<span class="sd"> (fname) will receive a given value (fval), given that the</span>
<span class="sd"> label (label).</span>
<span class="sd"> If the classifier encounters an input with a feature that has</span>
<span class="sd"> never been seen with any label, then rather than assigning a</span>
<span class="sd"> probability of 0 to all labels, it will ignore that feature.</span>
<span class="sd"> The feature value 'None' is reserved for unseen feature values;</span>
<span class="sd"> you generally should not use 'None' as a feature value for one of</span>
<span class="sd"> your own features.</span>
<span class="sd"> """</span>
<div class="viewcode-block" id="NaiveBayesClassifier.__init__"><a class="viewcode-back" href="../../../api/nltk.classify.naivebayes.html#nltk.classify.naivebayes.NaiveBayesClassifier.__init__">[docs]</a> <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">label_probdist</span><span class="p">,</span> <span class="n">feature_probdist</span><span class="p">):</span>
<span class="sd">"""</span>
<span class="sd"> :param label_probdist: P(label), the probability distribution</span>
<span class="sd"> over labels. It is expressed as a ``ProbDistI`` whose</span>
<span class="sd"> samples are labels. I.e., P(label) =</span>
<span class="sd"> ``label_probdist.prob(label)``.</span>
<span class="sd"> :param feature_probdist: P(fname=fval|label), the probability</span>
<span class="sd"> distribution for feature values, given labels. It is</span>
<span class="sd"> expressed as a dictionary whose keys are ``(label, fname)``</span>
<span class="sd"> pairs and whose values are ``ProbDistI`` objects over feature</span>
<span class="sd"> values. I.e., P(fname=fval|label) =</span>
<span class="sd"> ``feature_probdist[label,fname].prob(fval)``. If a given</span>
<span class="sd"> ``(label,fname)`` is not a key in ``feature_probdist``, then</span>
<span class="sd"> it is assumed that the corresponding P(fname=fval|label)</span>
<span class="sd"> is 0 for all values of ``fval``.</span>
<span class="sd"> """</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_label_probdist</span> <span class="o">=</span> <span class="n">label_probdist</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_feature_probdist</span> <span class="o">=</span> <span class="n">feature_probdist</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_labels</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">label_probdist</span><span class="o">.</span><span class="n">samples</span><span class="p">())</span></div>
<div class="viewcode-block" id="NaiveBayesClassifier.labels"><a class="viewcode-back" href="../../../api/nltk.classify.naivebayes.html#nltk.classify.naivebayes.NaiveBayesClassifier.labels">[docs]</a> <span class="k">def</span> <span class="nf">labels</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span></div>
<div class="viewcode-block" id="NaiveBayesClassifier.classify"><a class="viewcode-back" href="../../../api/nltk.classify.naivebayes.html#nltk.classify.naivebayes.NaiveBayesClassifier.classify">[docs]</a> <span class="k">def</span> <span class="nf">classify</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">featureset</span><span class="p">):</span>
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">prob_classify</span><span class="p">(</span><span class="n">featureset</span><span class="p">)</span><span class="o">.</span><span class="n">max</span><span class="p">()</span></div>
<div class="viewcode-block" id="NaiveBayesClassifier.prob_classify"><a class="viewcode-back" href="../../../api/nltk.classify.naivebayes.html#nltk.classify.naivebayes.NaiveBayesClassifier.prob_classify">[docs]</a> <span class="k">def</span> <span class="nf">prob_classify</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">featureset</span><span class="p">):</span>
<span class="c1"># Discard any feature names that we've never seen before.</span>
<span class="c1"># Otherwise, we'll just assign a probability of 0 to</span>
<span class="c1"># everything.</span>
<span class="n">featureset</span> <span class="o">=</span> <span class="n">featureset</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
<span class="k">for</span> <span class="n">fname</span> <span class="ow">in</span> <span class="nb">list</span><span class="p">(</span><span class="n">featureset</span><span class="o">.</span><span class="n">keys</span><span class="p">()):</span>
<span class="k">for</span> <span class="n">label</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="p">:</span>
<span class="k">if</span> <span class="p">(</span><span class="n">label</span><span class="p">,</span> <span class="n">fname</span><span class="p">)</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_feature_probdist</span><span class="p">:</span>
<span class="k">break</span>
<span class="k">else</span><span class="p">:</span>
<span class="c1"># print('Ignoring unseen feature %s' % fname)</span>
<span class="k">del</span> <span class="n">featureset</span><span class="p">[</span><span class="n">fname</span><span class="p">]</span>
<span class="c1"># Find the log probability of each label, given the features.</span>
<span class="c1"># Start with the log probability of the label itself.</span>
<span class="n">logprob</span> <span class="o">=</span> <span class="p">{}</span>
<span class="k">for</span> <span class="n">label</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="p">:</span>
<span class="n">logprob</span><span class="p">[</span><span class="n">label</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_label_probdist</span><span class="o">.</span><span class="n">logprob</span><span class="p">(</span><span class="n">label</span><span class="p">)</span>
<span class="c1"># Then add in the log probability of features given labels.</span>
<span class="k">for</span> <span class="n">label</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span><span class="p">:</span>
<span class="k">for</span> <span class="p">(</span><span class="n">fname</span><span class="p">,</span> <span class="n">fval</span><span class="p">)</span> <span class="ow">in</span> <span class="n">featureset</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
<span class="k">if</span> <span class="p">(</span><span class="n">label</span><span class="p">,</span> <span class="n">fname</span><span class="p">)</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_feature_probdist</span><span class="p">:</span>
<span class="n">feature_probs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_feature_probdist</span><span class="p">[</span><span class="n">label</span><span class="p">,</span> <span class="n">fname</span><span class="p">]</span>
<span class="n">logprob</span><span class="p">[</span><span class="n">label</span><span class="p">]</span> <span class="o">+=</span> <span class="n">feature_probs</span><span class="o">.</span><span class="n">logprob</span><span class="p">(</span><span class="n">fval</span><span class="p">)</span>
<span class="k">else</span><span class="p">:</span>
<span class="c1"># nb: This case will never come up if the</span>
<span class="c1"># classifier was created by</span>
<span class="c1"># NaiveBayesClassifier.train().</span>
<span class="n">logprob</span><span class="p">[</span><span class="n">label</span><span class="p">]</span> <span class="o">+=</span> <span class="n">sum_logs</span><span class="p">([])</span> <span class="c1"># = -INF.</span>
<span class="k">return</span> <span class="n">DictionaryProbDist</span><span class="p">(</span><span class="n">logprob</span><span class="p">,</span> <span class="n">normalize</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">log</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span></div>
<div class="viewcode-block" id="NaiveBayesClassifier.show_most_informative_features"><a class="viewcode-back" href="../../../api/nltk.classify.naivebayes.html#nltk.classify.naivebayes.NaiveBayesClassifier.show_most_informative_features">[docs]</a> <span class="k">def</span> <span class="nf">show_most_informative_features</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">n</span><span class="o">=</span><span class="mi">10</span><span class="p">):</span>
<span class="c1"># Determine the most relevant features, and display them.</span>
<span class="n">cpdist</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_feature_probdist</span>
<span class="nb">print</span><span class="p">(</span><span class="s2">"Most Informative Features"</span><span class="p">)</span>
<span class="k">for</span> <span class="p">(</span><span class="n">fname</span><span class="p">,</span> <span class="n">fval</span><span class="p">)</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">most_informative_features</span><span class="p">(</span><span class="n">n</span><span class="p">):</span>
<span class="k">def</span> <span class="nf">labelprob</span><span class="p">(</span><span class="n">l</span><span class="p">):</span>
<span class="k">return</span> <span class="n">cpdist</span><span class="p">[</span><span class="n">l</span><span class="p">,</span> <span class="n">fname</span><span class="p">]</span><span class="o">.</span><span class="n">prob</span><span class="p">(</span><span class="n">fval</span><span class="p">)</span>
<span class="n">labels</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span>
<span class="p">(</span><span class="n">l</span> <span class="k">for</span> <span class="n">l</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_labels</span> <span class="k">if</span> <span class="n">fval</span> <span class="ow">in</span> <span class="n">cpdist</span><span class="p">[</span><span class="n">l</span><span class="p">,</span> <span class="n">fname</span><span class="p">]</span><span class="o">.</span><span class="n">samples</span><span class="p">()),</span>
<span class="n">key</span><span class="o">=</span><span class="k">lambda</span> <span class="n">element</span><span class="p">:</span> <span class="p">(</span><span class="o">-</span><span class="n">labelprob</span><span class="p">(</span><span class="n">element</span><span class="p">),</span> <span class="n">element</span><span class="p">),</span>
<span class="n">reverse</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
<span class="p">)</span>
<span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">labels</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
<span class="k">continue</span>
<span class="n">l0</span> <span class="o">=</span> <span class="n">labels</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
<span class="n">l1</span> <span class="o">=</span> <span class="n">labels</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
<span class="k">if</span> <span class="n">cpdist</span><span class="p">[</span><span class="n">l0</span><span class="p">,</span> <span class="n">fname</span><span class="p">]</span><span class="o">.</span><span class="n">prob</span><span class="p">(</span><span class="n">fval</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
<span class="n">ratio</span> <span class="o">=</span> <span class="s2">"INF"</span>
<span class="k">else</span><span class="p">:</span>
<span class="n">ratio</span> <span class="o">=</span> <span class="s2">"</span><span class="si">%8.1f</span><span class="s2">"</span> <span class="o">%</span> <span class="p">(</span>
<span class="n">cpdist</span><span class="p">[</span><span class="n">l1</span><span class="p">,</span> <span class="n">fname</span><span class="p">]</span><span class="o">.</span><span class="n">prob</span><span class="p">(</span><span class="n">fval</span><span class="p">)</span> <span class="o">/</span> <span class="n">cpdist</span><span class="p">[</span><span class="n">l0</span><span class="p">,</span> <span class="n">fname</span><span class="p">]</span><span class="o">.</span><span class="n">prob</span><span class="p">(</span><span class="n">fval</span><span class="p">)</span>
<span class="p">)</span>
<span class="nb">print</span><span class="p">(</span>
<span class="s2">"</span><span class="si">%24s</span><span class="s2"> = </span><span class="si">%-14r</span><span class="s2"> </span><span class="si">%6s</span><span class="s2"> : </span><span class="si">%-6s</span><span class="s2"> = </span><span class="si">%s</span><span class="s2"> : 1.0"</span>
<span class="o">%</span> <span class="p">(</span><span class="n">fname</span><span class="p">,</span> <span class="n">fval</span><span class="p">,</span> <span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="n">l1</span><span class="p">)[:</span><span class="mi">6</span><span class="p">],</span> <span class="p">(</span><span class="s2">"</span><span class="si">%s</span><span class="s2">"</span> <span class="o">%</span> <span class="n">l0</span><span class="p">)[:</span><span class="mi">6</span><span class="p">],</span> <span class="n">ratio</span><span class="p">)</span>
<span class="p">)</span></div>
<div class="viewcode-block" id="NaiveBayesClassifier.most_informative_features"><a class="viewcode-back" href="../../../api/nltk.classify.naivebayes.html#nltk.classify.naivebayes.NaiveBayesClassifier.most_informative_features">[docs]</a> <span class="k">def</span> <span class="nf">most_informative_features</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">n</span><span class="o">=</span><span class="mi">100</span><span class="p">):</span>
<span class="sd">"""</span>
<span class="sd"> Return a list of the 'most informative' features used by this</span>
<span class="sd"> classifier. For the purpose of this function, the</span>
<span class="sd"> informativeness of a feature ``(fname,fval)`` is equal to the</span>
<span class="sd"> highest value of P(fname=fval|label), for any label, divided by</span>
<span class="sd"> the lowest value of P(fname=fval|label), for any label:</span>
<span class="sd"> | max[ P(fname=fval|label1) / P(fname=fval|label2) ]</span>
<span class="sd"> """</span>
<span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">"_most_informative_features"</span><span class="p">):</span>
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_most_informative_features</span><span class="p">[:</span><span class="n">n</span><span class="p">]</span>
<span class="k">else</span><span class="p">:</span>
<span class="c1"># The set of (fname, fval) pairs used by this classifier.</span>
<span class="n">features</span> <span class="o">=</span> <span class="nb">set</span><span class="p">()</span>
<span class="c1"># The max & min probability associated w/ each (fname, fval)</span>
<span class="c1"># pair. Maps (fname,fval) -> float.</span>
<span class="n">maxprob</span> <span class="o">=</span> <span class="n">defaultdict</span><span class="p">(</span><span class="k">lambda</span><span class="p">:</span> <span class="mf">0.0</span><span class="p">)</span>
<span class="n">minprob</span> <span class="o">=</span> <span class="n">defaultdict</span><span class="p">(</span><span class="k">lambda</span><span class="p">:</span> <span class="mf">1.0</span><span class="p">)</span>
<span class="k">for</span> <span class="p">(</span><span class="n">label</span><span class="p">,</span> <span class="n">fname</span><span class="p">),</span> <span class="n">probdist</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">_feature_probdist</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
<span class="k">for</span> <span class="n">fval</span> <span class="ow">in</span> <span class="n">probdist</span><span class="o">.</span><span class="n">samples</span><span class="p">():</span>
<span class="n">feature</span> <span class="o">=</span> <span class="p">(</span><span class="n">fname</span><span class="p">,</span> <span class="n">fval</span><span class="p">)</span>
<span class="n">features</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">feature</span><span class="p">)</span>
<span class="n">p</span> <span class="o">=</span> <span class="n">probdist</span><span class="o">.</span><span class="n">prob</span><span class="p">(</span><span class="n">fval</span><span class="p">)</span>
<span class="n">maxprob</span><span class="p">[</span><span class="n">feature</span><span class="p">]</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">maxprob</span><span class="p">[</span><span class="n">feature</span><span class="p">])</span>
<span class="n">minprob</span><span class="p">[</span><span class="n">feature</span><span class="p">]</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">minprob</span><span class="p">[</span><span class="n">feature</span><span class="p">])</span>
<span class="k">if</span> <span class="n">minprob</span><span class="p">[</span><span class="n">feature</span><span class="p">]</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
<span class="n">features</span><span class="o">.</span><span class="n">discard</span><span class="p">(</span><span class="n">feature</span><span class="p">)</span>
<span class="c1"># Convert features to a list, & sort it by how informative</span>
<span class="c1"># features are.</span>
<span class="bp">self</span><span class="o">.</span><span class="n">_most_informative_features</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span>
<span class="n">features</span><span class="p">,</span>
<span class="n">key</span><span class="o">=</span><span class="k">lambda</span> <span class="n">feature_</span><span class="p">:</span> <span class="p">(</span>
<span class="n">minprob</span><span class="p">[</span><span class="n">feature_</span><span class="p">]</span> <span class="o">/</span> <span class="n">maxprob</span><span class="p">[</span><span class="n">feature_</span><span class="p">],</span>
<span class="n">feature_</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
<span class="n">feature_</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="ow">in</span> <span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">False</span><span class="p">,</span> <span class="kc">True</span><span class="p">],</span>
<span class="nb">str</span><span class="p">(</span><span class="n">feature_</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">lower</span><span class="p">(),</span>
<span class="p">),</span>
<span class="p">)</span>
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_most_informative_features</span><span class="p">[:</span><span class="n">n</span><span class="p">]</span></div>
<div class="viewcode-block" id="NaiveBayesClassifier.train"><a class="viewcode-back" href="../../../api/nltk.classify.naivebayes.html#nltk.classify.naivebayes.NaiveBayesClassifier.train">[docs]</a> <span class="nd">@classmethod</span>
<span class="k">def</span> <span class="nf">train</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="n">labeled_featuresets</span><span class="p">,</span> <span class="n">estimator</span><span class="o">=</span><span class="n">ELEProbDist</span><span class="p">):</span>
<span class="sd">"""</span>
<span class="sd"> :param labeled_featuresets: A list of classified featuresets,</span>
<span class="sd"> i.e., a list of tuples ``(featureset, label)``.</span>
<span class="sd"> """</span>
<span class="n">label_freqdist</span> <span class="o">=</span> <span class="n">FreqDist</span><span class="p">()</span>
<span class="n">feature_freqdist</span> <span class="o">=</span> <span class="n">defaultdict</span><span class="p">(</span><span class="n">FreqDist</span><span class="p">)</span>
<span class="n">feature_values</span> <span class="o">=</span> <span class="n">defaultdict</span><span class="p">(</span><span class="nb">set</span><span class="p">)</span>
<span class="n">fnames</span> <span class="o">=</span> <span class="nb">set</span><span class="p">()</span>
<span class="c1"># Count up how many times each feature value occurred, given</span>
<span class="c1"># the label and featurename.</span>
<span class="k">for</span> <span class="n">featureset</span><span class="p">,</span> <span class="n">label</span> <span class="ow">in</span> <span class="n">labeled_featuresets</span><span class="p">:</span>
<span class="n">label_freqdist</span><span class="p">[</span><span class="n">label</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
<span class="k">for</span> <span class="n">fname</span><span class="p">,</span> <span class="n">fval</span> <span class="ow">in</span> <span class="n">featureset</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
<span class="c1"># Increment freq(fval|label, fname)</span>
<span class="n">feature_freqdist</span><span class="p">[</span><span class="n">label</span><span class="p">,</span> <span class="n">fname</span><span class="p">][</span><span class="n">fval</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
<span class="c1"># Record that fname can take the value fval.</span>
<span class="n">feature_values</span><span class="p">[</span><span class="n">fname</span><span class="p">]</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">fval</span><span class="p">)</span>
<span class="c1"># Keep a list of all feature names.</span>
<span class="n">fnames</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">fname</span><span class="p">)</span>
<span class="c1"># If a feature didn't have a value given for an instance, then</span>
<span class="c1"># we assume that it gets the implicit value 'None.' This loop</span>
<span class="c1"># counts up the number of 'missing' feature values for each</span>
<span class="c1"># (label,fname) pair, and increments the count of the fval</span>
<span class="c1"># 'None' by that amount.</span>
<span class="k">for</span> <span class="n">label</span> <span class="ow">in</span> <span class="n">label_freqdist</span><span class="p">:</span>
<span class="n">num_samples</span> <span class="o">=</span> <span class="n">label_freqdist</span><span class="p">[</span><span class="n">label</span><span class="p">]</span>
<span class="k">for</span> <span class="n">fname</span> <span class="ow">in</span> <span class="n">fnames</span><span class="p">:</span>
<span class="n">count</span> <span class="o">=</span> <span class="n">feature_freqdist</span><span class="p">[</span><span class="n">label</span><span class="p">,</span> <span class="n">fname</span><span class="p">]</span><span class="o">.</span><span class="n">N</span><span class="p">()</span>
<span class="c1"># Only add a None key when necessary, i.e. if there are</span>
<span class="c1"># any samples with feature 'fname' missing.</span>
<span class="k">if</span> <span class="n">num_samples</span> <span class="o">-</span> <span class="n">count</span> <span class="o">></span> <span class="mi">0</span><span class="p">:</span>
<span class="n">feature_freqdist</span><span class="p">[</span><span class="n">label</span><span class="p">,</span> <span class="n">fname</span><span class="p">][</span><span class="kc">None</span><span class="p">]</span> <span class="o">+=</span> <span class="n">num_samples</span> <span class="o">-</span> <span class="n">count</span>
<span class="n">feature_values</span><span class="p">[</span><span class="n">fname</span><span class="p">]</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="kc">None</span><span class="p">)</span>
<span class="c1"># Create the P(label) distribution</span>
<span class="n">label_probdist</span> <span class="o">=</span> <span class="n">estimator</span><span class="p">(</span><span class="n">label_freqdist</span><span class="p">)</span>
<span class="c1"># Create the P(fval|label, fname) distribution</span>
<span class="n">feature_probdist</span> <span class="o">=</span> <span class="p">{}</span>
<span class="k">for</span> <span class="p">((</span><span class="n">label</span><span class="p">,</span> <span class="n">fname</span><span class="p">),</span> <span class="n">freqdist</span><span class="p">)</span> <span class="ow">in</span> <span class="n">feature_freqdist</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
<span class="n">probdist</span> <span class="o">=</span> <span class="n">estimator</span><span class="p">(</span><span class="n">freqdist</span><span class="p">,</span> <span class="n">bins</span><span class="o">=</span><span class="nb">len</span><span class="p">(</span><span class="n">feature_values</span><span class="p">[</span><span class="n">fname</span><span class="p">]))</span>
<span class="n">feature_probdist</span><span class="p">[</span><span class="n">label</span><span class="p">,</span> <span class="n">fname</span><span class="p">]</span> <span class="o">=</span> <span class="n">probdist</span>
<span class="k">return</span> <span class="bp">cls</span><span class="p">(</span><span class="n">label_probdist</span><span class="p">,</span> <span class="n">feature_probdist</span><span class="p">)</span></div></div>
<span class="c1">##//////////////////////////////////////////////////////</span>
<span class="c1">## Demo</span>
<span class="c1">##//////////////////////////////////////////////////////</span>
<div class="viewcode-block" id="demo"><a class="viewcode-back" href="../../../api/nltk.classify.naivebayes.html#nltk.classify.naivebayes.demo">[docs]</a><span class="k">def</span> <span class="nf">demo</span><span class="p">():</span>
<span class="kn">from</span> <span class="nn">nltk.classify.util</span> <span class="kn">import</span> <span class="n">names_demo</span>
<span class="n">classifier</span> <span class="o">=</span> <span class="n">names_demo</span><span class="p">(</span><span class="n">NaiveBayesClassifier</span><span class="o">.</span><span class="n">train</span><span class="p">)</span>
<span class="n">classifier</span><span class="o">.</span><span class="n">show_most_informative_features</span><span class="p">()</span></div>
<span class="k">if</span> <span class="vm">__name__</span> <span class="o">==</span> <span class="s2">"__main__"</span><span class="p">:</span>
<span class="n">demo</span><span class="p">()</span>
</pre></div>
</div>
</div>
</div>
<footer>
<div id="footer-info">
<ul id="build-details">
<li class="footer-element">
<a href="https://github.com/nltk/nltk/tree/3.6.5">3.6.5</a>
</li>
<li class="footer-element">
Oct 11, 2021
</li>
</ul>
<div id="copyright">
© 2021, NLTK Project
</div>
<div id="credit">
created with <a href="http://sphinx-doc.org/">Sphinx</a> and <a href="https://github.com/tomaarsen/nltk_theme">NLTK Theme</a>
</div>
</div>
</footer>
</div>
</body>
</html>