Make NLTKTagger (averaged perceptron) the default tagger

Author: sloria

CHANGELOG.rst

@@ -7,6 +7,7 @@ Changelog
 Changes:
 
 - Compatible with nltk>=3.1. NLTK versions < 3.1 are no longer supported.
+- Change default tagger to NLTKTagger (uses NLTK's averaged perceptron tagger).
 - Tested on Python 3.5.
 
 Bug fixes:

tests/test_blob.py

@@ -404,11 +404,11 @@ def test_words(self):
             'better',
             'than',
             'implicit',
-            ]))
+        ]))
         short = tb.TextBlob("Just a bundle of words")
         assert_equal(short.words, tb.WordList([
             'Just', 'a', 'bundle', 'of', 'words'
-            ]))
+        ]))
 
     def test_words_includes_apostrophes_in_contractions(self):
         blob = tb.TextBlob("Let's test this.")
@@ -421,7 +421,7 @@ def test_pos_tags(self):
         blob = tb.TextBlob('Simple is better than complex. '
                             'Complex is better than complicated.')
         assert_equal(blob.pos_tags, [
-            ('Simple', 'JJ'),
+            ('Simple', 'NN'),
             ('is', 'VBZ'),
             ('better', 'JJR'),
             ('than', 'IN'),
@@ -431,7 +431,7 @@ def test_pos_tags(self):
             ('better', 'JJR'),
             ('than', 'IN'),
             ('complicated', 'VBN'),
-            ])
+        ])
 
     def test_tags(self):
         assert_equal(self.blob.tags, self.blob.pos_tags)
@@ -442,7 +442,6 @@ def test_tagging_nonascii(self):
         tags = b.tags
         assert_true(isinstance(tags[0][0], unicode))
 
-
     def test_pos_tags_includes_one_letter_articles(self):
         blob = tb.TextBlob("This is a sentence.")
         assert_equal(blob.pos_tags[2][0], 'a')
@@ -483,14 +482,13 @@ def test_can_get_subjectivity_and_polarity_with_different_analyzer(self):
 
     def test_pos_tagger_defaults_to_pattern(self):
         blob = tb.TextBlob("some text")
-        assert_true(isinstance(blob.pos_tagger, PatternTagger))
+        assert_true(isinstance(blob.pos_tagger, NLTKTagger))
 
     def test_pos_tagger_is_shared_among_instances(self):
         blob1 = tb.TextBlob("This is one sentence")
         blob2 = tb.TextBlob("This is another sentence.")
         assert_true(blob1.pos_tagger is blob2.pos_tagger)
 
-
     def test_can_use_different_pos_tagger(self):
         tagger = NLTKTagger()
         blob = tb.TextBlob("this is some text", pos_tagger=tagger)
@@ -972,7 +970,7 @@ def test_creates_blobs(self):
 
     def test_default_tagger(self):
         blob = self.blobber("Some text")
-        assert_true(isinstance(blob.pos_tagger, PatternTagger))
+        assert_true(isinstance(blob.pos_tagger, NLTKTagger))
 
     def test_default_np_extractor(self):
         blob = self.blobber("Some text")
@@ -983,7 +981,7 @@ def test_default_tokenizer(self):
         assert_true(isinstance(blob.tokenizer, WordTokenizer))
 
     def test_str_and_repr(self):
-        expected = "Blobber(tokenizer=WordTokenizer(), pos_tagger=PatternTagger(), np_extractor=FastNPExtractor(), analyzer=PatternAnalyzer(), parser=PatternParser(), classifier=None)"
+        expected = "Blobber(tokenizer=WordTokenizer(), pos_tagger=NLTKTagger(), np_extractor=FastNPExtractor(), analyzer=PatternAnalyzer(), parser=PatternParser(), classifier=None)"
         assert_equal(repr(self.blobber), expected)
         assert_equal(str(self.blobber), repr(self.blobber))
 

textblob/blob.py

@@ -35,7 +35,7 @@
 from textblob.base import (BaseNPExtractor, BaseTagger, BaseTokenizer,
                        BaseSentimentAnalyzer, BaseParser)
 from textblob.np_extractors import FastNPExtractor
-from textblob.taggers import PatternTagger
+from textblob.taggers import NLTKTagger
 from textblob.tokenizers import WordTokenizer, sent_tokenize, word_tokenize
 from textblob.sentiments import PatternAnalyzer
 from textblob.parsers import PatternParser
@@ -321,7 +321,7 @@ class BaseBlob(StringlikeMixin, BlobComparableMixin):
     :param np_extractor: (optional) An NPExtractor instance. If ``None``,
         defaults to :class:`FastNPExtractor() <textblob.en.np_extractors.FastNPExtractor>`.
     :param pos_tagger: (optional) A Tagger instance. If ``None``,
-        defaults to :class:`PatternTagger <textblob.en.taggers.PatternTagger>`.
+        defaults to :class:`NLTKTagger <textblob.en.taggers.NLTKTagger>`.
     :param analyzer: (optional) A sentiment analyzer. If ``None``,
         defaults to :class:`PatternAnalyzer <textblob.en.sentiments.PatternAnalyzer>`.
     :param parser: A parser. If ``None``, defaults to
@@ -332,7 +332,7 @@ class BaseBlob(StringlikeMixin, BlobComparableMixin):
         ``clean_html`` parameter deprecated, as it was in NLTK.
     """
     np_extractor = FastNPExtractor()
-    pos_tagger = PatternTagger()
+    pos_tagger = NLTKTagger()
     tokenizer = WordTokenizer()
     translator = Translator()
     analyzer = PatternAnalyzer()
@@ -589,7 +589,7 @@ class TextBlob(BaseBlob):
     :param np_extractor: (optional) An NPExtractor instance. If ``None``,
         defaults to :class:`FastNPExtractor() <textblob.en.np_extractors.FastNPExtractor>`.
     :param pos_tagger: (optional) A Tagger instance. If ``None``, defaults to
-        :class:`PatternTagger <textblob.en.taggers.PatternTagger>`.
+        :class:`NLTKTagger <textblob.en.taggers.NLTKTagger>`.
     :param analyzer: (optional) A sentiment analyzer. If ``None``, defaults to
         :class:`PatternAnalyzer <textblob.en.sentiments.PatternAnalyzer>`.
     :param classifier: (optional) A classifier.
@@ -711,7 +711,7 @@ class Blobber(object):
     :param np_extractor: (optional) An NPExtractor instance. If ``None``,
         defaults to :class:`FastNPExtractor() <textblob.en.np_extractors.FastNPExtractor>`.
     :param pos_tagger: (optional) A Tagger instance. If ``None``,
-        defaults to :class:`PatternTagger <textblob.en.taggers.PatternTagger>`.
+        defaults to :class:`NLTKTagger <textblob.en.taggers.NLTKTagger>`.
     :param analyzer: (optional) A sentiment analyzer. If ``None``,
         defaults to :class:`PatternAnalyzer <textblob.en.sentiments.PatternAnalyzer>`.
     :param parser: A parser. If ``None``, defaults to
@@ -722,7 +722,7 @@ class Blobber(object):
     """
 
     np_extractor = FastNPExtractor()
-    pos_tagger = PatternTagger()
+    pos_tagger = NLTKTagger()
     tokenizer = WordTokenizer()
     analyzer = PatternAnalyzer()
     parser = PatternParser()