Rooba
/
spaCy
mirror of https://github.com/explosion/spaCy.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Add missing we've and hardcoded 's and 'S

This commit is contained in:
jxs8172 2015-08-21 22:57:47 -04:00
parent a5e0a0073b
commit 5876248109
2 changed files with 7 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
lang_data/en/generate_specials.py
View File

@ -94,7 +94,7 @@ starting_tokens = {
"there": {"'d": [], "'d've": [], "'s": ["contrLower", "contrUpper"]},
"they": {"'d": [], "'d've": [], "'ll": [], "'re": [], "'ve": []},
"was": {"n't": []},
"we": {"'d": ["contrLower", "contrUpper"], "'d've": [], "'ll": ["contrLower", "contrUpper"], "'re": ["contrLower", "contrUpper"]},
"we": {"'d": ["contrLower", "contrUpper"], "'d've": [], "'ll": ["contrLower", "contrUpper"], "'re": ["contrLower", "contrUpper"], "'ve": []},
"were": {"n't": []},
"what": {"'ll": [], "'re": [], "'s": [], "'ve": []},
"when": {"'s": []},
@ -110,6 +110,11 @@ starting_tokens = {
# other specials that don't really have contractions
# so they are hardcoded
hardcoded_specials = {
"'s": [{"F": "'s", "L": "'s"}],
"'S": [{"F": "'S", "L": "'s"}],
"'em": [{"F": "'em"}],
"'ol": [{"F": "'ol"}],

2
lang_data/en/specials.json
View File

File diff suppressed because one or more lines are too long