explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-12-24 17:06:29 +03:00
Code Issues Packages Projects Releases Wiki Activity

Add support for Universal Dependencies v2.0

Browse Source
This commit is contained in:
Roman Inflianskas 2017-02-26 22:27:11 +01:00
parent 8dff040032
commit 66e1109b53
14 changed files with 155 additions and 74 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
spacy/attrs.pyx
View File

@ -125,7 +125,9 @@ def intify_attrs(stringy_attrs, strings_map=None, _do_deprecated=False):
'VerbForm', 'PronType', 'Aspect', 'Tense', 'PartType', 'Poss',
'Hyph', 'ConjType', 'NumType', 'Foreign', 'VerbType', 'NounType',
'Number', 'PronType', 'AdjType', 'Person', 'Variant', 'AdpType',
'Reflex', 'Negative', 'Mood', 'Aspect', 'Case']
'Reflex', 'Negative', 'Mood', 'Aspect', 'Case',
'Polarity', # U20
]
for key in morph_keys:
if key in stringy_attrs:
stringy_attrs.pop(key)

2
spacy/de/tag_map.py
View File

@ -41,7 +41,7 @@ TAG_MAP = {
"PRF": {POS: PRON, "PronType": "prs", "Reflex": "yes"},
"PTKA": {POS: PART},
"PTKANT": {POS: PART, "PartType": "res"},
"PTKNEG": {POS: PART, "Negative": "yes"},
"PTKNEG": {POS: PART, "Polarity": "Neg"},
"PTKVZ": {POS: PART, "PartType": "vbp"},
"PTKZU": {POS: PART, "PartType": "inf"},
"PWAT": {POS: DET, "PronType": "int"},

2
spacy/en/tag_map.py
View File

@ -16,7 +16,7 @@ TAG_MAP = {
"$": {POS: SYM, "Other": {"SymType": "currency"}},
"#": {POS: SYM, "Other": {"SymType": "numbersign"}},
"AFX": {POS: ADJ, "Hyph": "yes"},
"CC": {POS: CONJ, "ConjType": "coor"},
"CC": {POS: CCONJ, "ConjType": "coor"},
"CD": {POS: NUM, "NumType": "card"},
"DT": {POS: DET},
"EX": {POS: ADV, "AdvType": "ex"},

1
spacy/language_data/tag_map.py
View File

@ -19,6 +19,7 @@ TAG_MAP = {
"AUX": {POS: AUX},
"X": {POS: X},
"CONJ": {POS: CONJ},
"CCONJ": {POS: CCONJ}, # U20
"ADJ": {POS: ADJ},
"VERB": {POS: VERB},
"PART": {POS: PART}

4
spacy/morphology.pxd
View File

@ -80,6 +80,7 @@ cpdef enum univ_morph_t:
Definite_two
Definite_def
Definite_red
Definite_cons # U20
Definite_ind
Degree_cmp
Degree_comp
@ -103,6 +104,8 @@ cpdef enum univ_morph_t:
Negative_neg
Negative_pos
Negative_yes
Polarity_neg # U20
Polarity_pos # U20
Number_com
Number_dual
Number_none
@ -151,6 +154,7 @@ cpdef enum univ_morph_t:
VerbForm_partPres
VerbForm_sup
VerbForm_trans
VerbForm_conv # U20
VerbForm_gdv # la
Voice_act
Voice_cau

4
spacy/morphology.pyx
View File

@ -192,6 +192,7 @@ IDS = {
"Definite_two": Definite_two,
"Definite_def": Definite_def,
"Definite_red": Definite_red,
"Definite_cons": Definite_cons, # U20
"Definite_ind": Definite_ind,
"Degree_cmp": Degree_cmp,
"Degree_comp": Degree_comp,
@ -215,6 +216,8 @@ IDS = {
"Negative_neg": Negative_neg,
"Negative_pos": Negative_pos,
"Negative_yes": Negative_yes,
"Polarity_neg": Polarity_neg, # U20
"Polarity_pos": Polarity_pos, # U20
"Number_com": Number_com,
"Number_dual": Number_dual,
"Number_none": Number_none,
@ -263,6 +266,7 @@ IDS = {
"VerbForm_partPres": VerbForm_partPres,
"VerbForm_sup": VerbForm_sup,
"VerbForm_trans": VerbForm_trans,
"VerbForm_conv": VerbForm_conv, # U20
"VerbForm_gdv ": VerbForm_gdv, # la,
"Voice_act": Voice_act,
"Voice_cau": Voice_cau,

1
spacy/parts_of_speech.pxd
View File

@ -7,6 +7,7 @@ cpdef enum univ_pos_t:
ADV
AUX
CONJ
CCONJ # U20
DET
INTJ
NOUN

3
spacy/parts_of_speech.pyx
View File

@ -7,7 +7,8 @@ IDS = {
"ADP": ADP,
"ADV": ADV,
"AUX": AUX,
"CONJ": CONJ,
"CONJ": CONJ, # U20
"CCONJ": CCONJ,
"DET": DET,
"INTJ": INTJ,
"NOUN": NOUN,

40
spacy/symbols.pxd
View File

@ -90,6 +90,7 @@ cpdef enum symbol_t:
ADV
AUX
CONJ
CCONJ # U20
DET
INTJ
NOUN
@ -107,11 +108,14 @@ cpdef enum symbol_t:
Animacy_anim
Animacy_inam
Animacy_hum # U20
Aspect_freq
Aspect_imp
Aspect_mod
Aspect_none
Aspect_perf
Aspect_iter # U20
Aspect_hab # U20
Case_abe
Case_abl
Case_abs
@ -120,10 +124,12 @@ cpdef enum symbol_t:
Case_all
Case_cau
Case_com
Case_cmp # U20
Case_dat
Case_del
Case_dis
Case_ela
Case_equ # U20
Case_ess
Case_gen
Case_ill
@ -142,7 +148,9 @@ cpdef enum symbol_t:
Definite_two
Definite_def
Definite_red
Definite_cons # U20
Definite_ind
Definite_spec # U20
Degree_cmp
Degree_comp
Degree_none
@ -151,6 +159,8 @@ cpdef enum symbol_t:
Degree_abs
Degree_com
Degree_dim # du
Degree_equ # U20
Evident_nfh # U20
Gender_com
Gender_fem
Gender_masc
@ -162,16 +172,21 @@ cpdef enum symbol_t:
Mood_pot
Mood_sub
Mood_opt
Mood_prp # U20
Mood_adm # U20
Negative_neg
Negative_pos
Negative_yes
Polarity_neg # U20
Polarity_pos # U20
Number_com
Number_dual
Number_none
Number_plur
Number_sing
Number_ptan # bg
Number_count # bg
Number_count # bg, U20
Number_tri # U20
NumType_card
NumType_dist
NumType_frac
@ -197,7 +212,8 @@ cpdef enum symbol_t:
PronType_rel
PronType_tot
PronType_clit
PronType_exc # es, ca, it, fa
PronType_exc # es, ca, it, fa, U20
PronType_emp # U20
Reflex_yes
Tense_fut
Tense_imp
@ -213,12 +229,17 @@ cpdef enum symbol_t:
VerbForm_partPres
VerbForm_sup
VerbForm_trans
VerbForm_conv # U20
VerbForm_gdv # la
VerbForm_vnoun # U20
Voice_act
Voice_cau
Voice_pass
Voice_mid # gkc
Voice_mid # gkc, U20
Voice_int # hb
Voice_antip # U20
Voice_dir # U20
Voice_inv # U20
Abbr_yes # cz, fi, sl, U
AdpType_prep # cz, U
AdpType_post # U
@ -284,6 +305,10 @@ cpdef enum symbol_t:
Number_psee_plur # U
Number_psor_sing # cz, fi, sl, U
Number_psor_plur # cz, fi, sl, U
Number_pauc # U20
Number_grpa # U20
Number_grpl # U20
Number_inv # U20
NumForm_digit # cz, sl, U
NumForm_roman # cz, sl, U
NumForm_word # cz, sl, U
@ -311,6 +336,8 @@ cpdef enum symbol_t:
Person_psor_one # fi, U
Person_psor_two # fi, U
Person_psor_three # fi, U
Person_zero # U20
Person_four # U20
Polite_inf # bq, U
Polite_pol # bq, U
Polite_abs_inf # bq, U
@ -319,6 +346,10 @@ cpdef enum symbol_t:
Polite_erg_pol # bq, U
Polite_dat_inf # bq, U
Polite_dat_pol # bq, U
Polite_infm # U20
Polite_form # U20
Polite_form_elev # U20
Polite_form_humb # U20
Prefix_yes # U
PrepCase_npr # cz
PrepCase_pre # U
@ -383,6 +414,7 @@ cpdef enum symbol_t:
ccomp
complm
conj
cop # U20
csubj
csubjpass
dep
@ -405,6 +437,8 @@ cpdef enum symbol_t:
num
number
oprd
obj # U20
obl # U20
parataxis
partmod
pcomp

40
spacy/symbols.pyx
View File

@ -91,6 +91,7 @@ IDS = {
"ADV": ADV,
"AUX": AUX,
"CONJ": CONJ,
"CCONJ": CCONJ, # U20
"DET": DET,
"INTJ": INTJ,
"NOUN": NOUN,
@ -108,11 +109,14 @@ IDS = {
"Animacy_anim": Animacy_anim,
"Animacy_inam": Animacy_inam,
"Animacy_hum": Animacy_hum, # U20
"Aspect_freq": Aspect_freq,
"Aspect_imp": Aspect_imp,
"Aspect_mod": Aspect_mod,
"Aspect_none": Aspect_none,
"Aspect_perf": Aspect_perf,
"Aspect_iter": Aspect_iter, # U20
"Aspect_hab": Aspect_hab, # U20
"Case_abe": Case_abe,
"Case_abl": Case_abl,
"Case_abs": Case_abs,
@ -121,10 +125,12 @@ IDS = {
"Case_all": Case_all,
"Case_cau": Case_cau,
"Case_com": Case_com,
"Case_cmp": Case_cmp, # U20
"Case_dat": Case_dat,
"Case_del": Case_del,
"Case_dis": Case_dis,
"Case_ela": Case_ela,
"Case_equ": Case_equ, # U20
"Case_ess": Case_ess,
"Case_gen": Case_gen,
"Case_ill": Case_ill,
@ -143,7 +149,9 @@ IDS = {
"Definite_two": Definite_two,
"Definite_def": Definite_def,
"Definite_red": Definite_red,
"Definite_cons": Definite_cons, # U20
"Definite_ind": Definite_ind,
"Definite_spec": Definite_spec, # U20
"Degree_cmp": Degree_cmp,
"Degree_comp": Degree_comp,
"Degree_none": Degree_none,
@ -152,6 +160,8 @@ IDS = {
"Degree_abs": Degree_abs,
"Degree_com": Degree_com,
"Degree_dim ": Degree_dim, # du
"Degree_equ": Degree_equ, # U20
"Evident_nfh": Evident_nfh, # U20
"Gender_com": Gender_com,
"Gender_fem": Gender_fem,
"Gender_masc": Gender_masc,
@ -163,16 +173,21 @@ IDS = {
"Mood_pot": Mood_pot,
"Mood_sub": Mood_sub,
"Mood_opt": Mood_opt,
"Mood_prp": Mood_prp, # U20
"Mood_adm": Mood_adm, # U20
"Negative_neg": Negative_neg,
"Negative_pos": Negative_pos,
"Negative_yes": Negative_yes,
"Polarity_neg": Polarity_neg, # U20
"Polarity_pos": Polarity_pos, # U20
"Number_com": Number_com,
"Number_dual": Number_dual,
"Number_none": Number_none,
"Number_plur": Number_plur,
"Number_sing": Number_sing,
"Number_ptan ": Number_ptan, # bg
"Number_count ": Number_count, # bg
"Number_count ": Number_count, # bg, U20
"Number_tri": Number_tri, # U20
"NumType_card": NumType_card,
"NumType_dist": NumType_dist,
"NumType_frac": NumType_frac,
@ -198,7 +213,8 @@ IDS = {
"PronType_rel": PronType_rel,
"PronType_tot": PronType_tot,
"PronType_clit": PronType_clit,
"PronType_exc ": PronType_exc, # es, ca, it, fa,
"PronType_exc": PronType_exc, # es, ca, it, fa, U20
"PronType_emp": PronType_emp, # U20
"Reflex_yes": Reflex_yes,
"Tense_fut": Tense_fut,
"Tense_imp": Tense_imp,
@ -214,12 +230,17 @@ IDS = {
"VerbForm_partPres": VerbForm_partPres,
"VerbForm_sup": VerbForm_sup,
"VerbForm_trans": VerbForm_trans,
"VerbForm_conv": VerbForm_conv, # U20
"VerbForm_gdv ": VerbForm_gdv, # la,
"VerbForm_vnoun": VerbForm_vnoun, # U20
"Voice_act": Voice_act,
"Voice_cau": Voice_cau,
"Voice_pass": Voice_pass,
"Voice_mid ": Voice_mid, # gkc,
"Voice_mid ": Voice_mid, # gkc, U20
"Voice_int ": Voice_int, # hb,
"Voice_antip": Voice_antip, # U20
"Voice_dir": Voice_dir, # U20
"Voice_inv": Voice_inv, # U20
"Abbr_yes ": Abbr_yes, # cz, fi, sl, U,
"AdpType_prep ": AdpType_prep, # cz, U,
"AdpType_post ": AdpType_post, # U,
@ -285,6 +306,10 @@ IDS = {
"Number_psee_plur ": Number_psee_plur, # U,
"Number_psor_sing ": Number_psor_sing, # cz, fi, sl, U,
"Number_psor_plur ": Number_psor_plur, # cz, fi, sl, U,
"Number_pauc": Number_pauc, # U20
"Number_grpa": Number_grpa, # U20
"Number_grpl": Number_grpl, # U20
"Number_inv": Number_inv, # U20
"NumForm_digit ": NumForm_digit, # cz, sl, U,
"NumForm_roman ": NumForm_roman, # cz, sl, U,
"NumForm_word ": NumForm_word, # cz, sl, U,
@ -312,6 +337,8 @@ IDS = {
"Person_psor_one ": Person_psor_one, # fi, U,
"Person_psor_two ": Person_psor_two, # fi, U,
"Person_psor_three ": Person_psor_three, # fi, U,
"Person_zero ": Person_zero, # U20
"Person_four ": Person_four, # U20
"Polite_inf ": Polite_inf, # bq, U,
"Polite_pol ": Polite_pol, # bq, U,
"Polite_abs_inf ": Polite_abs_inf, # bq, U,
@ -320,6 +347,10 @@ IDS = {
"Polite_erg_pol ": Polite_erg_pol, # bq, U,
"Polite_dat_inf ": Polite_dat_inf, # bq, U,
"Polite_dat_pol ": Polite_dat_pol, # bq, U,
"Polite_infm ": Polite_infm, # U20
"Polite_form ": Polite_form, # U20
"Polite_form_elev ": Polite_form_elev, # U20
"Polite_form_humb ": Polite_form_humb, # U20
"Prefix_yes ": Prefix_yes, # U,
"PrepCase_npr ": PrepCase_npr, # cz,
"PrepCase_pre ": PrepCase_pre, # U,
@ -384,6 +415,7 @@ IDS = {
"ccomp": ccomp,
"complm": complm,
"conj": conj,
"cop": cop, # U20
"csubj": csubj,
"csubjpass": csubjpass,
"dep": dep,
@ -406,6 +438,8 @@ IDS = {
"num": num,
"number": number,
"oprd": oprd,
"obj": obj, # U20
"obl": obl, # U20
"parataxis": parataxis,
"partmod": partmod,
"pcomp": pcomp,

2
spacy/tagger.pyx
View File

@ -13,7 +13,7 @@ from thinc.linalg cimport VecVec
from .typedefs cimport attr_t
from .tokens.doc cimport Doc
from .attrs cimport TAG
from .parts_of_speech cimport NO_TAG, ADJ, ADV, ADP, CONJ, DET, NOUN, NUM, PRON
from .parts_of_speech cimport NO_TAG, ADJ, ADV, ADP, CCONJ, DET, NOUN, NUM, PRON
from .parts_of_speech cimport VERB, X, PUNCT, EOL, SPACE
from .gold cimport GoldParse

2
spacy/tokens/doc.pyx
View File

@ -16,7 +16,7 @@ from ..typedefs cimport attr_t, flags_t
from ..attrs cimport attr_id_t
from ..attrs cimport ID, ORTH, NORM, LOWER, SHAPE, PREFIX, SUFFIX, LENGTH, CLUSTER
from ..attrs cimport POS, LEMMA, TAG, DEP, HEAD, SPACY, ENT_IOB, ENT_TYPE
from ..parts_of_speech cimport CONJ, PUNCT, NOUN
from ..parts_of_speech cimport CCONJ, PUNCT, NOUN
from ..parts_of_speech cimport univ_pos_t
from ..lexeme cimport Lexeme
from .span cimport Span

2
spacy/tokens/token.pyx
View File

@ -20,7 +20,7 @@ from .. import parts_of_speech
from ..attrs cimport LEMMA
from ..attrs cimport ID, ORTH, NORM, LOWER, SHAPE, PREFIX, SUFFIX, LENGTH, CLUSTER
from ..attrs cimport POS, LEMMA, TAG, DEP
from ..parts_of_speech cimport CONJ, PUNCT
from ..parts_of_speech cimport CCONJ, PUNCT
from ..attrs cimport IS_ALPHA, IS_ASCII, IS_DIGIT, IS_LOWER, IS_PUNCT, IS_SPACE
from ..attrs cimport IS_BRACKET