explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-01-26 01:04:34 +03:00
Code Issues Packages Projects Releases Wiki Activity

* Fix lemma of let's, re Issue #177

Browse Source
This commit is contained in:
Matthew Honnibal 2015-11-13 06:42:23 +11:00
parent ec7d36c3a4
commit e8bd92f1e7
2 changed files with 14 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
lang_data/en/generate_specials.py
View File

@ -24,7 +24,6 @@ token_properties = {
"i": {"L": "-PRON-"}, # no POS for i? "i": {"L": "-PRON-"}, # no POS for i?
"is": {"L": "be", "pos": "VBZ"}, "is": {"L": "be", "pos": "VBZ"},
"it": {"L": "-PRON-"}, # no POS for it? "it": {"L": "-PRON-"}, # no POS for it?
"let": {}, # no POS or lemma for let?
"'ll": {"L": "will", "pos": "MD"}, "'ll": {"L": "will", "pos": "MD"},
"'m": {"L": "be", "pos": "VBP", "number": 1, "tenspect": 1}, "'m": {"L": "be", "pos": "VBP", "number": 1, "tenspect": 1},
"'ma": {}, # no POS or lemma for ma? "'ma": {}, # no POS or lemma for ma?
@ -83,7 +82,6 @@ starting_tokens = {
"i": {"'d": ["contrLower", "contrUpper"], "'d've": [], "'ll": ["contrLower", "contrUpper"], "'m": [], "'ma": [], "'ve": []}, "i": {"'d": ["contrLower", "contrUpper"], "'d've": [], "'ll": ["contrLower", "contrUpper"], "'m": [], "'ma": [], "'ve": []},
"is": {"n't": []}, "is": {"n't": []},
"it": {"'d": [], "'d've": [], "'ll": [], "'s": ["contrLower", "contrUpper"]}, "it": {"'d": [], "'d've": [], "'ll": [], "'s": ["contrLower", "contrUpper"]},
"let": {"'s": ["contrLower", "contrUpper"]},
"might": {"n't": [], "n't've": [], "'ve": []}, "might": {"n't": [], "n't've": [], "'ve": []},
"must": {"n't": [], "'ve": []}, "must": {"n't": [], "'ve": []},
"need": {"n't": []}, "need": {"n't": []},
@ -111,6 +109,8 @@ starting_tokens = {
# other specials that don't really have contractions # other specials that don't really have contractions
# so they are hardcoded # so they are hardcoded
hardcoded_specials = { hardcoded_specials = {
"let's": [{"F": "let"}, {"F": "'s", "L": "us"}],
"Let's": [{"F": "Let"}, {"F": "'s", "L": "us"}],
"'s": [{"F": "'s", "L": "'s"}], "'s": [{"F": "'s", "L": "'s"}],

22
lang_data/en/specials.json
View File

@ -2639,6 +2639,15 @@
"F": "ma" "F": "ma"
} }
], ],
"Let's": [
{
"F": "Let"
},
{
"L": "us",
"F": "'s"
}
],
"he's": [ "he's": [
{ {
"L": "-PRON-", "L": "-PRON-",
@ -3147,11 +3156,6 @@
"pos": "MD" "pos": "MD"
} }
], ],
"=)": [
{
"F": "=)"
}
],
"b.": [ "b.": [
{ {
"F": "b." "F": "b."
@ -3502,6 +3506,7 @@
"F": "let" "F": "let"
}, },
{ {
"L": "us",
"F": "'s" "F": "'s"
} }
], ],
@ -4531,12 +4536,9 @@
"pos": "VB" "pos": "VB"
} }
], ],
"Let's": [ "=)": [
{ {
"F": "Let" "F": "=)"
},
{
"F": "'s"
} }
], ],
"Colo.": [ "Colo.": [