Skip to content

Commit c791ca8

Browse files
committed
Feature: Add "calabaza" to ingredient_tagdict.json
1 parent 9e5df1f commit c791ca8

2 files changed

Lines changed: 6 additions & 0 deletions

File tree

docs/source/explanation/appendix/pos-tags.rst

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -37,3 +37,9 @@ This new ``tagdict`` contains over 1000 new words compared to NLTK's built-in ta
3737
ingredient_tagdict = load_ingredient_tagdict()
3838
tagger.tagdict.update(ingredient_tagdict)
3939
return _pos_tag(tokens=tokens, tagset=None, tagger=tagger, lang="eng")
40+
41+
.. note::
42+
43+
After initially generating the new ``tagdict``, we will manually insert additional entries for unambiguous words encountered in the wild that the tagger gets wrong, resulting in incorrect parsing of a sentence.
44+
45+
This means that the ``tagdict`` distributed with this library is not entirely reproducible using the approach described here.
74 Bytes
Binary file not shown.

0 commit comments

Comments
 (0)