explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-01-25 00:34:20 +03:00
Code Issues Packages Projects Releases Wiki Activity

* Remove em-dash from lemma rules. Handle instead in specials.

Browse Source
This commit is contained in:
Matthew Honnibal 2015-10-09 10:27:13 +02:00
commit d341443282
2 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
lang_data/en/generate_specials.py
View File

@ -1,3 +1,4 @@
# -#- coding: utf-8 -*-
import json import json
contractions = {"n't", "'nt", "not", "'ve", "'d", "'ll", "'s", "'m", "'ma", "'re"} contractions = {"n't", "'nt", "not", "'ve", "'d", "'ll", "'s", "'m", "'ma", "'re"}
@ -133,6 +134,8 @@ hardcoded_specials = {
"''": [{"F": "''"}], "''": [{"F": "''"}],
"—": [{"F": "—", "L": "--", "P": ":"}],
"Corp.": [{"F": "Corp."}], "Corp.": [{"F": "Corp."}],
"Inc.": [{"F": "Inc."}], "Inc.": [{"F": "Inc."}],
"Co.": [{"F": "Co."}], "Co.": [{"F": "Co."}],
@ -412,6 +415,6 @@ def generate_specials():
if __name__ == "__main__": if __name__ == "__main__":
specials = generate_specials() specials = generate_specials()
with open("specials.json", "w") as f: with open("specials.json", "w") as file_:
json.dump(specials, f) file_.write(json.dumps(specials, indent=2))

1
lang_data/en/lemma_rules.json
View File

@ -32,6 +32,5 @@
"punct": [ "punct": [
["“", "\""], ["“", "\""],
["”", "\""], ["”", "\""],
["—", "--"]
] ]
} }