mirror of
https://github.com/explosion/spaCy.git
synced 2024-12-24 17:06:29 +03:00
Merge pull request #5767 from adrianeboyd/feature/remove-tag-maps
This commit is contained in:
Ines Montani
GitHub
commit
cb65b36839
|
|
@ -131,8 +131,14 @@ def debug_data(
|
|||
tag_map = {}
|
||||
if tag_map_path is not None:
|
||||
tag_map = srsly.read_json(tag_map_path)
|
||||
morph_rules_path = util.ensure_path(config["training"]["morph_rules"])
|
||||
morph_rules = {}
|
||||
if morph_rules_path is not None:
|
||||
morph_rules = srsly.read_json(morph_rules_path)
|
||||
# Replace tag map with provided mapping
|
||||
nlp.vocab.morphology.load_tag_map(tag_map)
|
||||
# Load morph rules
|
||||
nlp.vocab.morphology.load_morph_exceptions(morph_rules)
|
||||
|
||||
msg.divider("Data file validation")
|
||||
|
||||
|
|
|
|||
|
|
@ -88,7 +88,7 @@ def train(
|
|||
require_gpu(use_gpu)
|
||||
else:
|
||||
msg.info("Using CPU")
|
||||
raw_text, tag_map, weights_data = load_from_paths(config)
|
||||
raw_text, tag_map, morph_rules, weights_data = load_from_paths(config)
|
||||
if config["training"]["seed"] is not None:
|
||||
fix_random_seed(config["training"]["seed"])
|
||||
if config["training"].get("use_pytorch_for_gpu_memory"):
|
||||
|
|
@ -127,6 +127,9 @@ def train(
|
|||
# Replace tag map with provided mapping
|
||||
nlp.vocab.morphology.load_tag_map(tag_map)
|
||||
|
||||
# Load morph rules
|
||||
nlp.vocab.morphology.load_morph_exceptions(morph_rules)
|
||||
|
||||
# Create empty extra lexeme tables so the data from spacy-lookups-data
|
||||
# isn't loaded if these features are accessed
|
||||
if config["training"]["omit_extra_lookups"]:
|
||||
|
|
@ -482,6 +485,12 @@ def load_from_paths(config):
|
|||
if not tag_map_path.exists():
|
||||
msg.fail("Can't find tag map path", tag_map_path, exits=1)
|
||||
tag_map = srsly.read_json(config["training"]["tag_map"])
|
||||
morph_rules = {}
|
||||
morph_rules_path = util.ensure_path(config["training"]["morph_rules"])
|
||||
if morph_rules_path is not None:
|
||||
if not morph_rules_path.exists():
|
||||
msg.fail("Can't find tag map path", morph_rules_path, exits=1)
|
||||
morph_rules = srsly.read_json(config["training"]["morph_rules"])
|
||||
weights_data = None
|
||||
init_tok2vec = util.ensure_path(config["training"]["init_tok2vec"])
|
||||
if init_tok2vec is not None:
|
||||
|
|
@ -489,7 +498,7 @@ def load_from_paths(config):
|
|||
msg.fail("Can't find pretrained tok2vec", init_tok2vec, exits=1)
|
||||
with init_tok2vec.open("rb") as file_:
|
||||
weights_data = file_.read()
|
||||
return raw_text, tag_map, weights_data
|
||||
return raw_text, tag_map, morph_rules, weights_data
|
||||
|
||||
|
||||
def verify_cli_args(
|
||||
|
|
|
|||
|
|
@ -1,6 +1,5 @@
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .punctuation import TOKENIZER_PREFIXES, TOKENIZER_SUFFIXES, TOKENIZER_INFIXES
|
||||
from .tag_map import TAG_MAP
|
||||
from .stop_words import STOP_WORDS
|
||||
|
||||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
|
|
@ -13,7 +12,6 @@ class BengaliDefaults(Language.Defaults):
|
|||
lex_attr_getters = dict(Language.Defaults.lex_attr_getters)
|
||||
lex_attr_getters[LANG] = lambda text: "bn"
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
tag_map = TAG_MAP
|
||||
stop_words = STOP_WORDS
|
||||
prefixes = TOKENIZER_PREFIXES
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
|
|
|
|||
|
|
@ -1,263 +0,0 @@
|
|||
from ...symbols import LEMMA, PRON_LEMMA
|
||||
|
||||
|
||||
MORPH_RULES = {
|
||||
"PRP": {
|
||||
"ঐ": {LEMMA: PRON_LEMMA, "PronType": "Dem"},
|
||||
"ওই": {LEMMA: PRON_LEMMA, "PronType": "Dem"},
|
||||
"আমাকে": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "One",
|
||||
"PronType": "Prs",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"কি": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
"PronType": "Int",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"সে": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "Three",
|
||||
"PronType": "Prs",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"কিসে": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
"PronType": "Int",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"তাকে": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "Three",
|
||||
"PronType": "Prs",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"স্বয়ং": {LEMMA: PRON_LEMMA, "Reflex": "Yes", "PronType": "Ref"},
|
||||
"কোনগুলো": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Gender": "Neut",
|
||||
"PronType": "Int",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"তুমি": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "Two",
|
||||
"PronType": "Prs",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"তুই": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "Two",
|
||||
"PronType": "Prs",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"তাদেরকে": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "Three",
|
||||
"PronType": "Prs",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"আমরা": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "One ",
|
||||
"PronType": "Prs",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"যিনি": {LEMMA: PRON_LEMMA, "Number": "Sing", "PronType": "Rel", "Case": "Nom"},
|
||||
"আমাদেরকে": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "One",
|
||||
"PronType": "Prs",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"কোন": {LEMMA: PRON_LEMMA, "Number": "Sing", "PronType": "Int", "Case": "Acc"},
|
||||
"কারা": {LEMMA: PRON_LEMMA, "Number": "Plur", "PronType": "Int", "Case": "Acc"},
|
||||
"তোমাকে": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "Two",
|
||||
"PronType": "Prs",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"তোকে": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "Two",
|
||||
"PronType": "Prs",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"খোদ": {LEMMA: PRON_LEMMA, "Reflex": "Yes", "PronType": "Ref"},
|
||||
"কে": {LEMMA: PRON_LEMMA, "Number": "Sing", "PronType": "Int", "Case": "Acc"},
|
||||
"যারা": {LEMMA: PRON_LEMMA, "Number": "Plur", "PronType": "Rel", "Case": "Nom"},
|
||||
"যে": {LEMMA: PRON_LEMMA, "Number": "Sing", "PronType": "Rel", "Case": "Nom"},
|
||||
"তোমরা": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "Two",
|
||||
"PronType": "Prs",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"তোরা": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "Two",
|
||||
"PronType": "Prs",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"তোমাদেরকে": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "Two",
|
||||
"PronType": "Prs",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"তোদেরকে": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "Two",
|
||||
"PronType": "Prs",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"আপন": {LEMMA: PRON_LEMMA, "Reflex": "Yes", "PronType": "Ref"},
|
||||
"এ": {LEMMA: PRON_LEMMA, "PronType": "Dem"},
|
||||
"নিজ": {LEMMA: PRON_LEMMA, "Reflex": "Yes", "PronType": "Ref"},
|
||||
"কার": {LEMMA: PRON_LEMMA, "Number": "Sing", "PronType": "Int", "Case": "Acc"},
|
||||
"যা": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
"PronType": "Rel",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"তারা": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "Three",
|
||||
"PronType": "Prs",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"আমি": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "One",
|
||||
"PronType": "Prs",
|
||||
"Case": "Nom",
|
||||
},
|
||||
},
|
||||
"PRP$": {
|
||||
"আমার": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "One",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"মোর": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "One",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"মোদের": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "One",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"তার": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "Three",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"তাহাার": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "Three",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"তোমাদের": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "Two",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"আমাদের": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "One",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"তোমার": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "Two",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"তোর": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Sing",
|
||||
"Person": "Two",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"তাদের": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "Three",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"কাদের": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"PronType": "Int",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"তোদের": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"Person": "Two",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"যাদের": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Number": "Plur",
|
||||
"PronType": "Int",
|
||||
"Case": "Acc",
|
||||
},
|
||||
},
|
||||
}
|
||||
|
|
@ -1,79 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, ADJ, CONJ, SCONJ, NUM, DET, ADV, ADP, X, VERB
|
||||
from ...symbols import CCONJ, NOUN, PROPN, PART, INTJ, SPACE, PRON, AUX, SYM
|
||||
|
||||
|
||||
TAG_MAP = {
|
||||
".": {POS: PUNCT, "PunctType": "peri"},
|
||||
",": {POS: PUNCT, "PunctType": "comm"},
|
||||
"-LRB-": {POS: PUNCT, "PunctType": "brck", "PunctSide": "ini"},
|
||||
"-RRB-": {POS: PUNCT, "PunctType": "brck", "PunctSide": "fin"},
|
||||
"``": {POS: PUNCT, "PunctType": "quot", "PunctSide": "ini"},
|
||||
'""': {POS: PUNCT, "PunctType": "quot", "PunctSide": "fin"},
|
||||
"''": {POS: PUNCT, "PunctType": "quot", "PunctSide": "fin"},
|
||||
":": {POS: PUNCT},
|
||||
"৳": {POS: SYM, "SymType": "currency"},
|
||||
"#": {POS: SYM, "SymType": "numbersign"},
|
||||
"AFX": {POS: ADJ, "Hyph": "yes"},
|
||||
"CC": {POS: CONJ, "ConjType": "coor"},
|
||||
"CD": {POS: NUM, "NumType": "card"},
|
||||
"DT": {POS: DET},
|
||||
"EX": {POS: ADV, "AdvType": "ex"},
|
||||
"FW": {POS: X, "Foreign": "yes"},
|
||||
"HYPH": {POS: PUNCT, "PunctType": "dash"},
|
||||
"IN": {POS: ADP},
|
||||
"JJ": {POS: ADJ, "Degree": "pos"},
|
||||
"JJR": {POS: ADJ, "Degree": "comp"},
|
||||
"JJS": {POS: ADJ, "Degree": "sup"},
|
||||
"LS": {POS: PUNCT, "NumType": "ord"},
|
||||
"MD": {POS: VERB, "VerbType": "mod"},
|
||||
"NIL": {POS: ""},
|
||||
"NN": {POS: NOUN, "Number": "sing"},
|
||||
"NNP": {POS: PROPN, "NounType": "prop", "Number": "sing"},
|
||||
"NNPS": {POS: PROPN, "NounType": "prop", "Number": "plur"},
|
||||
"NNS": {POS: NOUN, "Number": "plur"},
|
||||
"PDT": {POS: ADJ, "AdjType": "pdt", "PronType": "prn"},
|
||||
"POS": {POS: PART, "Poss": "yes"},
|
||||
"PRP": {POS: PRON, "PronType": "prs"},
|
||||
"PRP$": {POS: ADJ, "PronType": "prs", "Poss": "yes"},
|
||||
"RB": {POS: ADV, "Degree": "pos"},
|
||||
"RBR": {POS: ADV, "Degree": "comp"},
|
||||
"RBS": {POS: ADV, "Degree": "sup"},
|
||||
"RP": {POS: PART},
|
||||
"TO": {POS: PART, "PartType": "inf", "VerbForm": "inf"},
|
||||
"UH": {POS: INTJ},
|
||||
"VB": {POS: VERB, "VerbForm": "inf"},
|
||||
"VBD": {POS: VERB, "VerbForm": "fin", "Tense": "past"},
|
||||
"VBG": {POS: VERB, "VerbForm": "part", "Tense": "pres", "Aspect": "prog"},
|
||||
"VBN": {POS: VERB, "VerbForm": "part", "Tense": "past", "Aspect": "perf"},
|
||||
"VBP": {POS: VERB, "VerbForm": "fin", "Tense": "pres"},
|
||||
"VBZ": {
|
||||
POS: VERB,
|
||||
"VerbForm": "fin",
|
||||
"Tense": "pres",
|
||||
"Number": "sing",
|
||||
"Person": 3,
|
||||
},
|
||||
"WDT": {POS: ADJ, "PronType": "int|rel"},
|
||||
"WP": {POS: NOUN, "PronType": "int|rel"},
|
||||
"WP$": {POS: ADJ, "Poss": "yes", "PronType": "int|rel"},
|
||||
"WRB": {POS: ADV, "PronType": "int|rel"},
|
||||
"SP": {POS: SPACE},
|
||||
"ADV": {POS: ADV},
|
||||
"NOUN": {POS: NOUN},
|
||||
"ADP": {POS: ADP},
|
||||
"PRON": {POS: PRON},
|
||||
"SCONJ": {POS: SCONJ},
|
||||
"PROPN": {POS: PROPN},
|
||||
"DET": {POS: DET},
|
||||
"SYM": {POS: SYM},
|
||||
"INTJ": {POS: INTJ},
|
||||
"PUNCT": {POS: PUNCT},
|
||||
"NUM": {POS: NUM},
|
||||
"AUX": {POS: AUX},
|
||||
"X": {POS: X},
|
||||
"CONJ": {POS: CONJ},
|
||||
"CCONJ": {POS: CCONJ},
|
||||
"ADJ": {POS: ADJ},
|
||||
"VERB": {POS: VERB},
|
||||
"PART": {POS: PART},
|
||||
}
|
||||
|
|
@ -2,7 +2,6 @@ from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
|||
from .punctuation import TOKENIZER_INFIXES, TOKENIZER_SUFFIXES
|
||||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .morph_rules import MORPH_RULES
|
||||
|
||||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
from ...language import Language
|
||||
|
|
@ -15,7 +14,6 @@ class DanishDefaults(Language.Defaults):
|
|||
lex_attr_getters.update(LEX_ATTRS)
|
||||
lex_attr_getters[LANG] = lambda text: "da"
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
morph_rules = MORPH_RULES
|
||||
infixes = TOKENIZER_INFIXES
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
stop_words = STOP_WORDS
|
||||
|
|
|
|||
|
|
@ -1,308 +0,0 @@
|
|||
from ...symbols import LEMMA, PRON_LEMMA
|
||||
|
||||
# Source: Danish Universal Dependencies and http://fjern-uv.dk/pronom.php
|
||||
|
||||
# Note: The Danish Universal Dependencies specify Case=Acc for all instances
|
||||
# of "den"/"det" even when the case is in fact "Nom". In the rules below, Case
|
||||
# is left unspecified for "den" and "det".
|
||||
|
||||
MORPH_RULES = {
|
||||
"PRON": {
|
||||
"jeg": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Case": "Nom",
|
||||
"Gender": "Com",
|
||||
}, # Case=Nom|Gender=Com|Number=Sing|Person=1|PronType=Prs
|
||||
"mig": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
"Gender": "Com",
|
||||
}, # Case=Acc|Gender=Com|Number=Sing|Person=1|PronType=Prs
|
||||
"min": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Com",
|
||||
}, # Gender=Com|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs
|
||||
"mit": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Neut",
|
||||
}, # Gender=Neut|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs
|
||||
"vor": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Com",
|
||||
}, # Gender=Com|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs|Style=Form
|
||||
"vort": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Neut",
|
||||
}, # Gender=Neut|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs|Style=Form
|
||||
"du": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Case": "Nom",
|
||||
"Gender": "Com",
|
||||
}, # Case=Nom|Gender=Com|Number=Sing|Person=2|PronType=Prs
|
||||
"dig": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
"Gender": "Com",
|
||||
}, # Case=Acc|Gender=Com|Number=Sing|Person=2|PronType=Prs
|
||||
"din": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Com",
|
||||
}, # Gender=Com|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs
|
||||
"dit": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Neut",
|
||||
}, # Gender=Neut|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs
|
||||
"han": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Case": "Nom",
|
||||
"Gender": "Com",
|
||||
}, # Case=Nom|Gender=Com|Number=Sing|Person=3|PronType=Prs
|
||||
"hun": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Case": "Nom",
|
||||
"Gender": "Com",
|
||||
}, # Case=Nom|Gender=Com|Number=Sing|Person=3|PronType=Prs
|
||||
"den": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Com",
|
||||
}, # Case=Acc|Gender=Com|Number=Sing|Person=3|PronType=Prs, See note above.
|
||||
"det": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
}, # Case=Acc|Gender=Neut|Number=Sing|Person=3|PronType=Prs See note above.
|
||||
"ham": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
"Gender": "Com",
|
||||
}, # Case=Acc|Gender=Com|Number=Sing|Person=3|PronType=Prs
|
||||
"hende": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
"Gender": "Com",
|
||||
}, # Case=Acc|Gender=Com|Number=Sing|Person=3|PronType=Prs
|
||||
"sin": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Com",
|
||||
"Reflex": "Yes",
|
||||
}, # Gender=Com|Number=Sing|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs|Reflex=Yes
|
||||
"sit": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Neut",
|
||||
"Reflex": "Yes",
|
||||
}, # Gender=Neut|Number=Sing|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs|Reflex=Yes
|
||||
"vi": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Case": "Nom",
|
||||
"Gender": "Com",
|
||||
}, # Case=Nom|Gender=Com|Number=Plur|Person=1|PronType=Prs
|
||||
"os": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Case": "Acc",
|
||||
"Gender": "Com",
|
||||
}, # Case=Acc|Gender=Com|Number=Plur|Person=1|PronType=Prs
|
||||
"mine": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
}, # Number=Plur|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs
|
||||
"vore": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
}, # Number=Plur|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs|Style=Form
|
||||
"I": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Plur",
|
||||
"Case": "Nom",
|
||||
"Gender": "Com",
|
||||
}, # Case=Nom|Gender=Com|Number=Plur|Person=2|PronType=Prs
|
||||
"jer": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Plur",
|
||||
"Case": "Acc",
|
||||
"Gender": "Com",
|
||||
}, # Case=Acc|Gender=Com|Number=Plur|Person=2|PronType=Prs
|
||||
"dine": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
}, # Number=Plur|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs
|
||||
"de": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Case": "Nom",
|
||||
}, # Case=Nom|Number=Plur|Person=3|PronType=Prs
|
||||
"dem": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Case": "Acc",
|
||||
}, # Case=Acc|Number=Plur|Person=3|PronType=Prs
|
||||
"sine": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
}, # Number=Plur|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs|Reflex=Yes
|
||||
"vores": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Poss": "Yes",
|
||||
}, # Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs
|
||||
"De": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Case": "Nom",
|
||||
"Gender": "Com",
|
||||
}, # Case=Nom|Gender=Com|Person=2|Polite=Form|PronType=Prs
|
||||
"Dem": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Case": "Acc",
|
||||
"Gender": "Com",
|
||||
}, # Case=Acc|Gender=Com|Person=2|Polite=Form|PronType=Prs
|
||||
"Deres": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Poss": "Yes",
|
||||
}, # Person=2|Polite=Form|Poss=Yes|PronType=Prs
|
||||
"jeres": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Poss": "Yes",
|
||||
}, # Number[psor]=Plur|Person=2|Poss=Yes|PronType=Prs
|
||||
"sig": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Case": "Acc",
|
||||
"Reflex": "Yes",
|
||||
}, # Case=Acc|Person=3|PronType=Prs|Reflex=Yes
|
||||
"hans": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Poss": "Yes",
|
||||
}, # Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs
|
||||
"hendes": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Poss": "Yes",
|
||||
}, # Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs
|
||||
"dens": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Poss": "Yes",
|
||||
}, # Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs
|
||||
"dets": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Poss": "Yes",
|
||||
}, # Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs
|
||||
"deres": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Poss": "Yes",
|
||||
}, # Number[psor]=Plur|Person=3|Poss=Yes|PronType=Prs
|
||||
},
|
||||
"VERB": {
|
||||
"er": {LEMMA: "være", "VerbForm": "Fin", "Tense": "Pres"},
|
||||
"var": {LEMMA: "være", "VerbForm": "Fin", "Tense": "Past"},
|
||||
},
|
||||
}
|
||||
|
||||
for tag, rules in MORPH_RULES.items():
|
||||
for key, attrs in dict(rules).items():
|
||||
rules[key.title()] = attrs
|
||||
|
|
@ -1,7 +1,6 @@
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .punctuation import TOKENIZER_PREFIXES, TOKENIZER_SUFFIXES
|
||||
from .punctuation import TOKENIZER_INFIXES
|
||||
from .tag_map import TAG_MAP
|
||||
from .stop_words import STOP_WORDS
|
||||
from .syntax_iterators import SYNTAX_ITERATORS
|
||||
|
||||
|
|
@ -18,7 +17,6 @@ class GermanDefaults(Language.Defaults):
|
|||
prefixes = TOKENIZER_PREFIXES
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
infixes = TOKENIZER_INFIXES
|
||||
tag_map = TAG_MAP
|
||||
stop_words = STOP_WORDS
|
||||
syntax_iterators = SYNTAX_ITERATORS
|
||||
single_orth_variants = [
|
||||
|
|
|
|||
|
|
@ -1,63 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, ADJ, CCONJ, SCONJ, NUM, DET, ADV, ADP, X
|
||||
from ...symbols import NOUN, PROPN, PART, INTJ, SPACE, PRON, AUX, VERB
|
||||
|
||||
|
||||
TAG_MAP = {
|
||||
"$(": {POS: PUNCT, "PunctType": "brck"},
|
||||
"$,": {POS: PUNCT, "PunctType": "comm"},
|
||||
"$.": {POS: PUNCT, "PunctType": "peri"},
|
||||
"ADJA": {POS: ADJ},
|
||||
"ADJD": {POS: ADJ},
|
||||
"ADV": {POS: ADV},
|
||||
"APPO": {POS: ADP, "AdpType": "post"},
|
||||
"APPR": {POS: ADP, "AdpType": "prep"},
|
||||
"APPRART": {POS: ADP, "AdpType": "prep", "PronType": "art"},
|
||||
"APZR": {POS: ADP, "AdpType": "circ"},
|
||||
"ART": {POS: DET, "PronType": "art"},
|
||||
"CARD": {POS: NUM, "NumType": "card"},
|
||||
"FM": {POS: X, "Foreign": "yes"},
|
||||
"ITJ": {POS: INTJ},
|
||||
"KOKOM": {POS: CCONJ, "ConjType": "comp"},
|
||||
"KON": {POS: CCONJ},
|
||||
"KOUI": {POS: SCONJ},
|
||||
"KOUS": {POS: SCONJ},
|
||||
"NE": {POS: PROPN},
|
||||
"NNE": {POS: PROPN},
|
||||
"NN": {POS: NOUN},
|
||||
"PAV": {POS: ADV, "PronType": "dem"},
|
||||
"PROAV": {POS: ADV, "PronType": "dem"},
|
||||
"PDAT": {POS: DET, "PronType": "dem"},
|
||||
"PDS": {POS: PRON, "PronType": "dem"},
|
||||
"PIAT": {POS: DET, "PronType": "ind|neg|tot"},
|
||||
"PIDAT": {POS: DET, "PronType": "ind|neg|tot"},
|
||||
"PIS": {POS: PRON, "PronType": "ind|neg|tot"},
|
||||
"PPER": {POS: PRON, "PronType": "prs"},
|
||||
"PPOSAT": {POS: DET, "Poss": "yes", "PronType": "prs"},
|
||||
"PPOSS": {POS: PRON, "Poss": "yes", "PronType": "prs"},
|
||||
"PRELAT": {POS: DET, "PronType": "rel"},
|
||||
"PRELS": {POS: PRON, "PronType": "rel"},
|
||||
"PRF": {POS: PRON, "PronType": "prs", "Reflex": "yes"},
|
||||
"PTKA": {POS: PART},
|
||||
"PTKANT": {POS: PART, "PartType": "res"},
|
||||
"PTKNEG": {POS: PART, "Polarity": "neg"},
|
||||
"PTKVZ": {POS: ADP, "PartType": "vbp"},
|
||||
"PTKZU": {POS: PART, "PartType": "inf"},
|
||||
"PWAT": {POS: DET, "PronType": "int"},
|
||||
"PWAV": {POS: ADV, "PronType": "int"},
|
||||
"PWS": {POS: PRON, "PronType": "int"},
|
||||
"TRUNC": {POS: X, "Hyph": "yes"},
|
||||
"VAFIN": {POS: AUX, "Mood": "ind", "VerbForm": "fin"},
|
||||
"VAIMP": {POS: AUX, "Mood": "imp", "VerbForm": "fin"},
|
||||
"VAINF": {POS: AUX, "VerbForm": "inf"},
|
||||
"VAPP": {POS: AUX, "Aspect": "perf", "VerbForm": "part"},
|
||||
"VMFIN": {POS: VERB, "Mood": "ind", "VerbForm": "fin", "VerbType": "mod"},
|
||||
"VMINF": {POS: VERB, "VerbForm": "inf", "VerbType": "mod"},
|
||||
"VMPP": {POS: VERB, "Aspect": "perf", "VerbForm": "part", "VerbType": "mod"},
|
||||
"VVFIN": {POS: VERB, "Mood": "ind", "VerbForm": "fin"},
|
||||
"VVIMP": {POS: VERB, "Mood": "imp", "VerbForm": "fin"},
|
||||
"VVINF": {POS: VERB, "VerbForm": "inf"},
|
||||
"VVIZU": {POS: VERB, "VerbForm": "inf"},
|
||||
"VVPP": {POS: VERB, "Aspect": "perf", "VerbForm": "part"},
|
||||
"XY": {POS: X},
|
||||
"_SP": {POS: SPACE},
|
||||
}
|
||||
|
|
@ -1,5 +1,4 @@
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from ..tag_map import TAG_MAP
|
||||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .lemmatizer import GreekLemmatizer
|
||||
|
|
@ -18,7 +17,6 @@ class GreekDefaults(Language.Defaults):
|
|||
lex_attr_getters[LANG] = lambda text: "el"
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
stop_words = STOP_WORDS
|
||||
tag_map = TAG_MAP
|
||||
prefixes = TOKENIZER_PREFIXES
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
infixes = TOKENIZER_INFIXES
|
||||
|
|
|
|||
|
|
@ -1,8 +1,6 @@
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .tag_map import TAG_MAP
|
||||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .morph_rules import MORPH_RULES
|
||||
from .syntax_iterators import SYNTAX_ITERATORS
|
||||
|
||||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
|
|
@ -20,9 +18,7 @@ class EnglishDefaults(Language.Defaults):
|
|||
lex_attr_getters.update(LEX_ATTRS)
|
||||
lex_attr_getters[LANG] = _return_en
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
tag_map = TAG_MAP
|
||||
stop_words = STOP_WORDS
|
||||
morph_rules = MORPH_RULES
|
||||
syntax_iterators = SYNTAX_ITERATORS
|
||||
single_orth_variants = [
|
||||
{"tags": ["NFP"], "variants": ["…", "..."]},
|
||||
|
|
|
|||
|
|
@ -1,490 +0,0 @@
|
|||
from ...symbols import LEMMA, PRON_LEMMA
|
||||
|
||||
# Several entries here look pretty suspicious. These will get the POS SCONJ
|
||||
# given the tag IN, when an adpositional reading seems much more likely for
|
||||
# a lot of these prepositions. I'm not sure what I was running in 04395ffa4
|
||||
# when I did this? It doesn't seem right.
|
||||
_subordinating_conjunctions = [
|
||||
"that",
|
||||
"if",
|
||||
"as",
|
||||
"because",
|
||||
# "of",
|
||||
# "for",
|
||||
# "before",
|
||||
# "in",
|
||||
"while",
|
||||
# "after",
|
||||
"since",
|
||||
"like",
|
||||
# "with",
|
||||
"so",
|
||||
# "to",
|
||||
# "by",
|
||||
# "on",
|
||||
# "about",
|
||||
"than",
|
||||
"whether",
|
||||
"although",
|
||||
# "from",
|
||||
"though",
|
||||
# "until",
|
||||
"unless",
|
||||
"once",
|
||||
# "without",
|
||||
# "at",
|
||||
# "into",
|
||||
"cause",
|
||||
# "over",
|
||||
"upon",
|
||||
"till",
|
||||
"whereas",
|
||||
# "beyond",
|
||||
"whilst",
|
||||
"except",
|
||||
"despite",
|
||||
"wether",
|
||||
# "then",
|
||||
"but",
|
||||
"becuse",
|
||||
"whie",
|
||||
# "below",
|
||||
# "against",
|
||||
"it",
|
||||
"w/out",
|
||||
# "toward",
|
||||
"albeit",
|
||||
"save",
|
||||
"besides",
|
||||
"becouse",
|
||||
"coz",
|
||||
"til",
|
||||
"ask",
|
||||
"i'd",
|
||||
"out",
|
||||
"near",
|
||||
"seince",
|
||||
# "towards",
|
||||
"tho",
|
||||
"sice",
|
||||
"will",
|
||||
]
|
||||
|
||||
# This seems kind of wrong too?
|
||||
# _relative_pronouns = ["this", "that", "those", "these"]
|
||||
|
||||
MORPH_RULES = {
|
||||
# "DT": {word: {"POS": "PRON"} for word in _relative_pronouns},
|
||||
"IN": {word: {"POS": "SCONJ"} for word in _subordinating_conjunctions},
|
||||
"NN": {
|
||||
"something": {"POS": "PRON"},
|
||||
"anyone": {"POS": "PRON"},
|
||||
"anything": {"POS": "PRON"},
|
||||
"nothing": {"POS": "PRON"},
|
||||
"someone": {"POS": "PRON"},
|
||||
"everything": {"POS": "PRON"},
|
||||
"everyone": {"POS": "PRON"},
|
||||
"everybody": {"POS": "PRON"},
|
||||
"nobody": {"POS": "PRON"},
|
||||
"somebody": {"POS": "PRON"},
|
||||
"anybody": {"POS": "PRON"},
|
||||
"any1": {"POS": "PRON"},
|
||||
},
|
||||
"PRP": {
|
||||
"I": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"me": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"you": {LEMMA: PRON_LEMMA, "POS": "PRON", "PronType": "Prs", "Person": "Two"},
|
||||
"he": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"him": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"she": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Fem",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"her": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Fem",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"it": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
"we": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"us": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"they": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"them": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"mine": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"his": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"hers": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Fem",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"its": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"ours": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"yours": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"theirs": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"myself": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"yourself": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Case": "Acc",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"himself": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
"Gender": "Masc",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"herself": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
"Gender": "Fem",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"itself": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
"Gender": "Neut",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"themself": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"ourselves": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Case": "Acc",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"yourselves": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Case": "Acc",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"themselves": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"POS": "PRON",
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Case": "Acc",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
},
|
||||
"PRP$": {
|
||||
"my": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
},
|
||||
"your": {LEMMA: PRON_LEMMA, "Person": "Two", "PronType": "Prs", "Poss": "Yes"},
|
||||
"his": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
},
|
||||
"her": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Fem",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
},
|
||||
"its": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
},
|
||||
"our": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
},
|
||||
"their": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"PronType": "Prs",
|
||||
"Poss": "Yes",
|
||||
},
|
||||
},
|
||||
"RB": {word: {"POS": "PART"} for word in ["not", "n't", "nt", "n’t"]},
|
||||
"VB": {
|
||||
word: {"POS": "AUX"}
|
||||
for word in ["be", "have", "do", "get", "of", "am", "are", "'ve"]
|
||||
},
|
||||
"VBN": {"been": {LEMMA: "be", "POS": "AUX"}},
|
||||
"VBG": {"being": {LEMMA: "be", "POS": "AUX"}},
|
||||
"VBZ": {
|
||||
"am": {
|
||||
LEMMA: "be",
|
||||
"POS": "AUX",
|
||||
"VerbForm": "Fin",
|
||||
"Person": "One",
|
||||
"Tense": "Pres",
|
||||
"Mood": "Ind",
|
||||
},
|
||||
"are": {
|
||||
LEMMA: "be",
|
||||
"POS": "AUX",
|
||||
"VerbForm": "Fin",
|
||||
"Person": "Two",
|
||||
"Tense": "Pres",
|
||||
"Mood": "Ind",
|
||||
},
|
||||
"is": {
|
||||
LEMMA: "be",
|
||||
"POS": "AUX",
|
||||
"VerbForm": "Fin",
|
||||
"Person": "Three",
|
||||
"Tense": "Pres",
|
||||
"Mood": "Ind",
|
||||
},
|
||||
"'re": {
|
||||
LEMMA: "be",
|
||||
"POS": "AUX",
|
||||
"VerbForm": "Fin",
|
||||
"Person": "Two",
|
||||
"Tense": "Pres",
|
||||
"Mood": "Ind",
|
||||
},
|
||||
"'s": {
|
||||
LEMMA: "be",
|
||||
"POS": "AUX",
|
||||
"VerbForm": "Fin",
|
||||
"Person": "Three",
|
||||
"Tense": "Pres",
|
||||
"Mood": "Ind",
|
||||
},
|
||||
"has": {LEMMA: "have", "POS": "AUX"},
|
||||
"does": {LEMMA: "do", "POS": "AUX"},
|
||||
},
|
||||
"VBP": {
|
||||
"are": {
|
||||
LEMMA: "be",
|
||||
"POS": "AUX",
|
||||
"VerbForm": "Fin",
|
||||
"Tense": "Pres",
|
||||
"Mood": "Ind",
|
||||
},
|
||||
"'re": {
|
||||
LEMMA: "be",
|
||||
"POS": "AUX",
|
||||
"VerbForm": "Fin",
|
||||
"Tense": "Pres",
|
||||
"Mood": "Ind",
|
||||
},
|
||||
"am": {
|
||||
LEMMA: "be",
|
||||
"POS": "AUX",
|
||||
"VerbForm": "Fin",
|
||||
"Person": "One",
|
||||
"Tense": "Pres",
|
||||
"Mood": "Ind",
|
||||
},
|
||||
"do": {"POS": "AUX"},
|
||||
"have": {"POS": "AUX"},
|
||||
"'m": {"POS": "AUX", LEMMA: "be"},
|
||||
"'ve": {"POS": "AUX"},
|
||||
"'s": {"POS": "AUX"},
|
||||
"is": {"POS": "AUX"},
|
||||
"'d": {"POS": "AUX"},
|
||||
},
|
||||
"VBD": {
|
||||
"was": {
|
||||
LEMMA: "be",
|
||||
"POS": "AUX",
|
||||
"VerbForm": "Fin",
|
||||
"Tense": "Past",
|
||||
"Number": "Sing",
|
||||
},
|
||||
"were": {
|
||||
LEMMA: "be",
|
||||
"POS": "AUX",
|
||||
"VerbForm": "Fin",
|
||||
"Tense": "Past",
|
||||
"Number": "Plur",
|
||||
},
|
||||
"did": {LEMMA: "do", "POS": "AUX"},
|
||||
"had": {LEMMA: "have", "POS": "AUX"},
|
||||
"'d": {LEMMA: "have", "POS": "AUX"},
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
for tag, rules in MORPH_RULES.items():
|
||||
for key, attrs in dict(rules).items():
|
||||
rules[key.title()] = attrs
|
||||
|
|
@ -1,69 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, SYM, ADJ, CCONJ, NUM, DET, ADV, ADP, X, VERB
|
||||
from ...symbols import NOUN, PROPN, PART, INTJ, SPACE, PRON
|
||||
|
||||
|
||||
TAG_MAP = {
|
||||
".": {POS: PUNCT, "PunctType": "peri"},
|
||||
",": {POS: PUNCT, "PunctType": "comm"},
|
||||
"-LRB-": {POS: PUNCT, "PunctType": "brck", "PunctSide": "ini"},
|
||||
"-RRB-": {POS: PUNCT, "PunctType": "brck", "PunctSide": "fin"},
|
||||
"``": {POS: PUNCT, "PunctType": "quot", "PunctSide": "ini"},
|
||||
'""': {POS: PUNCT, "PunctType": "quot", "PunctSide": "fin"},
|
||||
"''": {POS: PUNCT, "PunctType": "quot", "PunctSide": "fin"},
|
||||
":": {POS: PUNCT},
|
||||
"$": {POS: SYM},
|
||||
"#": {POS: SYM},
|
||||
"AFX": {POS: ADJ, "Hyph": "yes"},
|
||||
"CC": {POS: CCONJ, "ConjType": "comp"},
|
||||
"CD": {POS: NUM, "NumType": "card"},
|
||||
"DT": {POS: DET},
|
||||
"EX": {POS: PRON, "AdvType": "ex"},
|
||||
"FW": {POS: X, "Foreign": "yes"},
|
||||
"HYPH": {POS: PUNCT, "PunctType": "dash"},
|
||||
"IN": {POS: ADP},
|
||||
"JJ": {POS: ADJ, "Degree": "pos"},
|
||||
"JJR": {POS: ADJ, "Degree": "comp"},
|
||||
"JJS": {POS: ADJ, "Degree": "sup"},
|
||||
"LS": {POS: X, "NumType": "ord"},
|
||||
"MD": {POS: VERB, "VerbType": "mod"},
|
||||
"NIL": {POS: X},
|
||||
"NN": {POS: NOUN, "Number": "sing"},
|
||||
"NNP": {POS: PROPN, "NounType": "prop", "Number": "sing"},
|
||||
"NNPS": {POS: PROPN, "NounType": "prop", "Number": "plur"},
|
||||
"NNS": {POS: NOUN, "Number": "plur"},
|
||||
"PDT": {POS: DET},
|
||||
"POS": {POS: PART, "Poss": "yes"},
|
||||
"PRP": {POS: PRON, "PronType": "prs"},
|
||||
"PRP$": {POS: DET, "PronType": "prs", "Poss": "yes"},
|
||||
"RB": {POS: ADV, "Degree": "pos"},
|
||||
"RBR": {POS: ADV, "Degree": "comp"},
|
||||
"RBS": {POS: ADV, "Degree": "sup"},
|
||||
"RP": {POS: ADP},
|
||||
"SP": {POS: SPACE},
|
||||
"SYM": {POS: SYM},
|
||||
"TO": {POS: PART, "PartType": "inf", "VerbForm": "inf"},
|
||||
"UH": {POS: INTJ},
|
||||
"VB": {POS: VERB, "VerbForm": "inf"},
|
||||
"VBD": {POS: VERB, "VerbForm": "fin", "Tense": "past"},
|
||||
"VBG": {POS: VERB, "VerbForm": "part", "Tense": "pres", "Aspect": "prog"},
|
||||
"VBN": {POS: VERB, "VerbForm": "part", "Tense": "past", "Aspect": "perf"},
|
||||
"VBP": {POS: VERB, "VerbForm": "fin", "Tense": "pres"},
|
||||
"VBZ": {
|
||||
POS: VERB,
|
||||
"VerbForm": "fin",
|
||||
"Tense": "pres",
|
||||
"Number": "sing",
|
||||
"Person": "three",
|
||||
},
|
||||
"WDT": {POS: DET},
|
||||
"WP": {POS: PRON},
|
||||
"WP$": {POS: DET, "Poss": "yes"},
|
||||
"WRB": {POS: ADV},
|
||||
"ADD": {POS: X},
|
||||
"NFP": {POS: PUNCT},
|
||||
"GW": {POS: X},
|
||||
"XX": {POS: X},
|
||||
"BES": {POS: VERB},
|
||||
"HVS": {POS: VERB},
|
||||
"_SP": {POS: SPACE},
|
||||
}
|
||||
|
|
@ -1,5 +1,4 @@
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .tag_map import TAG_MAP
|
||||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .syntax_iterators import SYNTAX_ITERATORS
|
||||
|
|
@ -20,7 +19,6 @@ class SpanishDefaults(Language.Defaults):
|
|||
Language.Defaults.lex_attr_getters[NORM], BASE_NORMS
|
||||
)
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
tag_map = TAG_MAP
|
||||
infixes = TOKENIZER_INFIXES
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
stop_words = STOP_WORDS
|
||||
|
|
|
|||
|
|
@ -1,310 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, SYM, ADJ, NUM, DET, ADV, ADP, X, VERB
|
||||
from ...symbols import NOUN, PROPN, PART, INTJ, SPACE, PRON, SCONJ, AUX, CONJ
|
||||
|
||||
# fmt: off
|
||||
TAG_MAP = {
|
||||
"ADJ___": {"morph": "_", POS: ADJ},
|
||||
"ADJ__AdpType=Prep": {"morph": "AdpType=Prep", POS: ADJ},
|
||||
"ADJ__AdpType=Preppron|Gender=Masc|Number=Sing": {"morph": "AdpType=Preppron|Gender=Masc|Number=Sing", POS: ADV},
|
||||
"ADJ__AdvType=Tim": {"morph": "AdvType=Tim", POS: ADJ},
|
||||
"ADJ__Gender=Fem|Number=Plur": {"morph": "Gender=Fem|Number=Plur", POS: ADJ},
|
||||
"ADJ__Gender=Fem|Number=Plur|NumType=Ord": {"morph": "Gender=Fem|Number=Plur|NumType=Ord", POS: ADJ},
|
||||
"ADJ__Gender=Fem|Number=Plur|VerbForm=Part": {"morph": "Gender=Fem|Number=Plur|VerbForm=Part", POS: ADJ},
|
||||
"ADJ__Gender=Fem|Number=Sing": {"morph": "Gender=Fem|Number=Sing", POS: ADJ},
|
||||
"ADJ__Gender=Fem|Number=Sing|NumType=Ord": {"morph": "Gender=Fem|Number=Sing|NumType=Ord", POS: ADJ},
|
||||
"ADJ__Gender=Fem|Number=Sing|VerbForm=Part": {"morph": "Gender=Fem|Number=Sing|VerbForm=Part", POS: ADJ},
|
||||
"ADJ__Gender=Masc": {"morph": "Gender=Masc", POS: ADJ},
|
||||
"ADJ__Gender=Masc|Number=Plur": {"morph": "Gender=Masc|Number=Plur", POS: ADJ},
|
||||
"ADJ__Gender=Masc|Number=Plur|NumType=Ord": {"morph": "Gender=Masc|Number=Plur|NumType=Ord", POS: ADJ},
|
||||
"ADJ__Gender=Masc|Number=Plur|VerbForm=Part": {"morph": "Gender=Masc|Number=Plur|VerbForm=Part", POS: ADJ},
|
||||
"ADJ__Gender=Masc|Number=Sing": {"morph": "Gender=Masc|Number=Sing", POS: ADJ},
|
||||
"ADJ__Gender=Masc|Number=Sing|NumType=Ord": {"morph": "Gender=Masc|Number=Sing|NumType=Ord", POS: ADJ},
|
||||
"ADJ__Gender=Masc|Number=Sing|VerbForm=Part": {"morph": "Gender=Masc|Number=Sing|VerbForm=Part", POS: ADJ},
|
||||
"ADJ__Number=Plur": {"morph": "Number=Plur", POS: ADJ},
|
||||
"ADJ__Number=Sing": {"morph": "Number=Sing", POS: ADJ},
|
||||
"ADP__AdpType=Prep": {"morph": "AdpType=Prep", POS: ADP},
|
||||
"ADP__AdpType=Preppron|Gender=Fem|Number=Sing": {"morph": "AdpType=Preppron|Gender=Fem|Number=Sing", POS: ADP},
|
||||
"ADP__AdpType=Preppron|Gender=Masc|Number=Plur": {"morph": "AdpType=Preppron|Gender=Masc|Number=Plur", POS: ADP},
|
||||
"ADP__AdpType=Preppron|Gender=Masc|Number=Sing": {"morph": "AdpType=Preppron|Gender=Masc|Number=Sing", POS: ADP},
|
||||
"ADP": {POS: ADP},
|
||||
"ADV___": {"morph": "_", POS: ADV},
|
||||
"ADV__AdpType=Prep": {"morph": "AdpType=Prep", POS: ADV},
|
||||
"ADV__AdpType=Preppron|Gender=Masc|Number=Sing": {"morph": "AdpType=Preppron|Gender=Masc|Number=Sing", POS: ADV},
|
||||
"ADV__AdvType=Tim": {"morph": "AdvType=Tim", POS: ADV},
|
||||
"ADV__Gender=Masc|Number=Sing": {"morph": "Gender=Masc|Number=Sing", POS: ADV},
|
||||
"ADV__Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin", POS: ADV},
|
||||
"ADV__Negative=Neg": {"morph": "Negative=Neg", POS: ADV},
|
||||
"ADV__Number=Plur": {"morph": "Number=Plur", POS: ADV},
|
||||
"ADV__Polarity=Neg": {"morph": "Polarity=Neg", POS: ADV},
|
||||
"AUX__Gender=Fem|Number=Plur|Tense=Past|VerbForm=Part": {"morph": "Gender=Fem|Number=Plur|Tense=Past|VerbForm=Part", POS: AUX},
|
||||
"AUX__Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part": {"morph": "Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part", POS: AUX},
|
||||
"AUX__Gender=Masc|Number=Plur|Tense=Past|VerbForm=Part": {"morph": "Gender=Masc|Number=Plur|Tense=Past|VerbForm=Part", POS: AUX},
|
||||
"AUX__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part": {"morph": "Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part", POS: AUX},
|
||||
"AUX__Mood=Cnd|Number=Plur|Person=1|VerbForm=Fin": {"morph": "Mood=Cnd|Number=Plur|Person=1|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Cnd|Number=Plur|Person=3|VerbForm=Fin": {"morph": "Mood=Cnd|Number=Plur|Person=3|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Cnd|Number=Sing|Person=1|VerbForm=Fin": {"morph": "Mood=Cnd|Number=Sing|Person=1|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Cnd|Number=Sing|Person=2|VerbForm=Fin": {"morph": "Mood=Cnd|Number=Sing|Person=2|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Cnd|Number=Sing|Person=3|VerbForm=Fin": {"morph": "Mood=Cnd|Number=Sing|Person=3|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Imp|Number=Plur|Person=3|VerbForm=Fin": {"morph": "Mood=Imp|Number=Plur|Person=3|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Imp|Number=Sing|Person=2|VerbForm=Fin": {"morph": "Mood=Imp|Number=Sing|Person=2|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Imp|Number=Sing|Person=3|VerbForm=Fin": {"morph": "Mood=Imp|Number=Sing|Person=3|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=1|Tense=Past|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=1|Tense=Past|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=3|Tense=Past|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=3|Tense=Past|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=1|Tense=Fut|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=1|Tense=Fut|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=1|Tense=Past|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=1|Tense=Past|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=2|Tense=Fut|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=2|Tense=Fut|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=2|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=2|Tense=Imp|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=3|Tense=Past|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=3|Tense=Past|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Sub|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Sub|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Sub|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Sub|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Sub|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Sub|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Sub|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Sub|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Sub|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin", POS: AUX},
|
||||
"AUX__VerbForm=Ger": {"morph": "VerbForm=Ger", POS: AUX},
|
||||
"AUX__VerbForm=Inf": {"morph": "VerbForm=Inf", POS: AUX},
|
||||
"CCONJ___": {"morph": "_", POS: CONJ},
|
||||
"CONJ___": {"morph": "_", POS: CONJ},
|
||||
"DET__Definite=Def|Gender=Fem|Number=Plur|PronType=Art": {"morph": "Definite=Def|Gender=Fem|Number=Plur|PronType=Art", POS: DET},
|
||||
"DET__Definite=Def|Gender=Fem|Number=Sing|PronType=Art": {"morph": "Definite=Def|Gender=Fem|Number=Sing|PronType=Art", POS: DET},
|
||||
"DET__Definite=Def|Gender=Masc|Number=Plur|PronType=Art": {"morph": "Definite=Def|Gender=Masc|Number=Plur|PronType=Art", POS: DET},
|
||||
"DET__Definite=Def|Gender=Masc|Number=Sing|PronType=Art": {"morph": "Definite=Def|Gender=Masc|Number=Sing|PronType=Art", POS: DET},
|
||||
"DET__Definite=Def|Gender=Masc|PronType=Art": {"morph": "Definite=Def|Gender=Masc|PronType=Art", POS: DET},
|
||||
"DET__Definite=Def|Number=Sing|PronType=Art": {"morph": "Definite=Def|Number=Sing|PronType=Art", POS: DET},
|
||||
"DET__Definite=Ind|Gender=Fem|Number=Plur|PronType=Art": {"morph": "Definite=Ind|Gender=Fem|Number=Plur|PronType=Art", POS: DET},
|
||||
"DET__Definite=Ind|Gender=Fem|Number=Sing|NumType=Card|PronType=Art": {"morph": "Definite=Ind|Gender=Fem|Number=Sing|NumType=Card|PronType=Art", POS: DET},
|
||||
"DET__Definite=Ind|Gender=Fem|Number=Sing|PronType=Art": {"morph": "Definite=Ind|Gender=Fem|Number=Sing|PronType=Art", POS: DET},
|
||||
"DET__Definite=Ind|Gender=Masc|Number=Plur|PronType=Art": {"morph": "Definite=Ind|Gender=Masc|Number=Plur|PronType=Art", POS: DET},
|
||||
"DET__Definite=Ind|Gender=Masc|Number=Sing|NumType=Card|PronType=Art": {"morph": "Definite=Ind|Gender=Masc|Number=Sing|NumType=Card|PronType=Art", POS: DET},
|
||||
"DET__Definite=Ind|Gender=Masc|Number=Sing|PronType=Art": {"morph": "Definite=Ind|Gender=Masc|Number=Sing|PronType=Art", POS: DET},
|
||||
"DET__Gender=Fem|Number=Plur|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Plur|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Gender=Fem|Number=Plur|Number[psor]=Plur|Person=2|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Plur|Number[psor]=Plur|Person=2|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Gender=Fem|Number=Plur|Person=3|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Plur|Person=3|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Gender=Fem|Number=Plur|PronType=Art": {"morph": "Gender=Fem|Number=Plur|PronType=Art", POS: DET},
|
||||
"DET__Gender=Fem|Number=Plur|PronType=Dem": {"morph": "Gender=Fem|Number=Plur|PronType=Dem", POS: DET},
|
||||
"DET__Gender=Fem|Number=Plur|PronType=Ind": {"morph": "Gender=Fem|Number=Plur|PronType=Ind", POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing|Number[psor]=Plur|Person=2|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Sing|Number[psor]=Plur|Person=2|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing|Person=3|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Sing|Person=3|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Art": {"morph": "Gender=Fem|Number=Sing|PronType=Art", POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Dem": {"morph": "Gender=Fem|Number=Sing|PronType=Dem", POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Ind": {"morph": "Gender=Fem|Number=Sing|PronType=Ind", POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Int": {"morph": "Gender=Fem|Number=Sing|PronType=Int", POS: DET},
|
||||
"DET__Gender=Masc|Number=Plur|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs": {"morph": "Gender=Masc|Number=Plur|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Gender=Masc|Number=Plur|Person=3|Poss=Yes|PronType=Prs": {"morph": "Gender=Masc|Number=Plur|Person=3|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Gender=Masc|Number=Plur|PronType=Art": {"morph": "Gender=Masc|Number=Plur|PronType=Art", POS: DET},
|
||||
"DET__Gender=Masc|Number=Plur|PronType=Dem": {"morph": "Gender=Masc|Number=Plur|PronType=Dem", POS: DET},
|
||||
"DET__Gender=Masc|Number=Plur|PronType=Ind": {"morph": "Gender=Masc|Number=Plur|PronType=Ind", POS: DET},
|
||||
"DET__Gender=Masc|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs": {"morph": "Gender=Masc|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Gender=Masc|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs": {"morph": "Gender=Masc|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Gender=Masc|Number=Sing|Person=3|Poss=Yes|PronType=Prs": {"morph": "Gender=Masc|Number=Sing|Person=3|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Art": {"morph": "Gender=Masc|Number=Sing|PronType=Art", POS: DET},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Dem": {"morph": "Gender=Masc|Number=Sing|PronType=Dem", POS: DET},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Ind": {"morph": "Gender=Masc|Number=Sing|PronType=Ind", POS: DET},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Int": {"morph": "Gender=Masc|Number=Sing|PronType=Int", POS: DET},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Tot": {"morph": "Gender=Masc|Number=Sing|PronType=Tot", POS: DET},
|
||||
"DET__Number=Plur|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs": {"morph": "Number=Plur|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Number=Plur|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs": {"morph": "Number=Plur|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Number=Plur|Person=3|Poss=Yes|PronType=Prs": {"morph": "Number=Plur|Person=3|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Number=Plur|PronType=Dem": {"morph": "Number=Plur|PronType=Dem", POS: DET},
|
||||
"DET__Number=Plur|PronType=Ind": {"morph": "Number=Plur|PronType=Ind", POS: DET},
|
||||
"DET__Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs": {"morph": "Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs": {"morph": "Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Number=Sing|Person=3|Poss=Yes|PronType=Prs": {"morph": "Number=Sing|Person=3|Poss=Yes|PronType=Prs", POS: DET},
|
||||
"DET__Number=Sing|PronType=Dem": {"morph": "Number=Sing|PronType=Dem", POS: DET},
|
||||
"DET__Number=Sing|PronType=Ind": {"morph": "Number=Sing|PronType=Ind", POS: DET},
|
||||
"DET__PronType=Int": {"morph": "PronType=Int", POS: DET},
|
||||
"DET__PronType=Rel": {"morph": "PronType=Rel", POS: DET},
|
||||
"DET": {POS: DET},
|
||||
"INTJ___": {"morph": "_", POS: INTJ},
|
||||
"NOUN___": {"morph": "_", POS: NOUN},
|
||||
"NOUN__AdvType=Tim": {"morph": "AdvType=Tim", POS: NOUN},
|
||||
"NOUN__AdvType=Tim|Gender=Masc|Number=Sing": {"morph": "AdvType=Tim|Gender=Masc|Number=Sing", POS: NOUN},
|
||||
"NOUN__Gender=Fem": {"morph": "Gender=Fem", POS: NOUN},
|
||||
"NOUN__Gender=Fem|Number=Plur": {"morph": "Gender=Fem|Number=Plur", POS: NOUN},
|
||||
"NOUN__Gender=Fem|Number=Sing": {"morph": "Gender=Fem|Number=Sing", POS: NOUN},
|
||||
"NOUN__Gender=Masc": {"morph": "Gender=Masc", POS: NOUN},
|
||||
"NOUN__Gender=Masc|Number=Plur": {"morph": "Gender=Masc|Number=Plur", POS: NOUN},
|
||||
"NOUN__Gender=Masc|Number=Sing": {"morph": "Gender=Masc|Number=Sing", POS: NOUN},
|
||||
"NOUN__Gender=Masc|Number=Sing|VerbForm=Part": {"morph": "Gender=Masc|Number=Sing|VerbForm=Part", POS: NOUN},
|
||||
"NOUN__Number=Plur": {"morph": "Number=Plur", POS: NOUN},
|
||||
"NOUN__Number=Sing": {"morph": "Number=Sing", POS: NOUN},
|
||||
"NOUN__NumForm=Digit": {"morph": "NumForm=Digit", POS: NOUN},
|
||||
"NUM__Gender=Fem|Number=Plur|NumType=Card": {"morph": "Gender=Fem|Number=Plur|NumType=Card", POS: NUM},
|
||||
"NUM__Gender=Fem|Number=Sing|NumType=Card": {"morph": "Gender=Fem|Number=Sing|NumType=Card", POS: NUM},
|
||||
"NUM__Gender=Masc|Number=Plur|NumType=Card": {"morph": "Gender=Masc|Number=Plur|NumType=Card", POS: NUM},
|
||||
"NUM__Gender=Masc|Number=Sing|NumType=Card": {"morph": "Gender=Masc|Number=Sing|NumType=Card", POS: NUM},
|
||||
"NUM__Number=Plur|NumType=Card": {"morph": "Number=Plur|NumType=Card", POS: NUM},
|
||||
"NUM__Number=Sing|NumType=Card": {"morph": "Number=Sing|NumType=Card", POS: NUM},
|
||||
"NUM__NumForm=Digit": {"morph": "NumForm=Digit", POS: NUM},
|
||||
"NUM__NumForm=Digit|NumType=Card": {"morph": "NumForm=Digit|NumType=Card", POS: NUM},
|
||||
"NUM__NumForm=Digit|NumType=Frac": {"morph": "NumForm=Digit|NumType=Frac", POS: NUM},
|
||||
"NUM__NumType=Card": {"morph": "NumType=Card", POS: NUM},
|
||||
"PART___": {"morph": "_", POS: PART},
|
||||
"PART__Negative=Neg": {"morph": "Negative=Neg", POS: PART},
|
||||
"PRON___": {"morph": "_", POS: PRON},
|
||||
"PRON__Case=Acc|Gender=Fem|Number=Plur|Person=3|PronType=Prs": {"morph": "Case=Acc|Gender=Fem|Number=Plur|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Case=Acc|Gender=Fem|Number=Sing|Person=3|PronType=Prs": {"morph": "Case=Acc|Gender=Fem|Number=Sing|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Case=Acc|Gender=Masc|Number=Plur|Person=3|PronType=Prs": {"morph": "Case=Acc|Gender=Masc|Number=Plur|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Case=Acc|Gender=Masc|Number=Sing|Person=3|PronType=Prs": {"morph": "Case=Acc|Gender=Masc|Number=Sing|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Case=Acc|Number=Plur|Person=3|PronType=Prs": {"morph": "Case=Acc|Number=Plur|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Case=Acc|Number=Sing|Person=3|PronType=Prs": {"morph": "Case=Acc|Number=Sing|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Case=Acc|Person=3|PronType=Prs": {"morph": "Case=Acc|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Case=Dat|Number=Plur|Person=3|PronType=Prs": {"morph": "Case=Dat|Number=Plur|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Case=Dat|Number=Sing|Person=3|PronType=Prs": {"morph": "Case=Dat|Number=Sing|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Case=Nom|Number=Sing|Person=1|PronType=Prs": {"morph": "Case=Nom|Number=Sing|Person=1|PronType=Prs", POS: PRON},
|
||||
"PRON__Case=Nom|Number=Sing|Person=2|PronType=Prs": {"morph": "Case=Nom|Number=Sing|Person=2|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Plur|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Plur|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Plur|Person=3|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Plur|Person=3|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Plur|Person=3|PronType=Prs": {"morph": "Gender=Fem|Number=Plur|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Plur|PronType=Dem": {"morph": "Gender=Fem|Number=Plur|PronType=Dem", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Plur|PronType=Ind": {"morph": "Gender=Fem|Number=Plur|PronType=Ind", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Plur|PronType=Int": {"morph": "Gender=Fem|Number=Plur|PronType=Int", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Plur|PronType=Rel": {"morph": "Gender=Fem|Number=Plur|PronType=Rel", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|Person=1|PronType=Prs": {"morph": "Gender=Fem|Number=Sing|Person=1|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|Person=3|Poss=Yes|PronType=Prs": {"morph": "Gender=Fem|Number=Sing|Person=3|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|Person=3|PronType=Prs": {"morph": "Gender=Fem|Number=Sing|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|PronType=Dem": {"morph": "Gender=Fem|Number=Sing|PronType=Dem", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|PronType=Ind": {"morph": "Gender=Fem|Number=Sing|PronType=Ind", POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|PronType=Rel": {"morph": "Gender=Fem|Number=Sing|PronType=Rel", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur|Person=1|PronType=Prs": {"morph": "Gender=Masc|Number=Plur|Person=1|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur|Person=2|Poss=Yes|PronType=Prs": {"morph": "Gender=Masc|Number=Plur|Person=2|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur|Person=3|PronType=Prs": {"morph": "Gender=Masc|Number=Plur|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur|PronType=Dem": {"morph": "Gender=Masc|Number=Plur|PronType=Dem", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur|PronType=Ind": {"morph": "Gender=Masc|Number=Plur|PronType=Ind", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur|PronType=Int": {"morph": "Gender=Masc|Number=Plur|PronType=Int", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur|PronType=Rel": {"morph": "Gender=Masc|Number=Plur|PronType=Rel", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs": {"morph": "Gender=Masc|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs": {"morph": "Gender=Masc|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs": {"morph": "Gender=Masc|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|Person=3|Poss=Yes|PronType=Prs": {"morph": "Gender=Masc|Number=Sing|Person=3|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|Person=3|PronType=Prs": {"morph": "Gender=Masc|Number=Sing|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|PronType=Dem": {"morph": "Gender=Masc|Number=Sing|PronType=Dem", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|PronType=Ind": {"morph": "Gender=Masc|Number=Sing|PronType=Ind", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|PronType=Int": {"morph": "Gender=Masc|Number=Sing|PronType=Int", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|PronType=Rel": {"morph": "Gender=Masc|Number=Sing|PronType=Rel", POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|PronType=Tot": {"morph": "Gender=Masc|Number=Sing|PronType=Tot", POS: PRON},
|
||||
"PRON__Number=Plur|Person=1": {"morph": "Number=Plur|Person=1", POS: PRON},
|
||||
"PRON__Number=Plur|Person=1|PronType=Prs": {"morph": "Number=Plur|Person=1|PronType=Prs", POS: PRON},
|
||||
"PRON__Number=Plur|Person=2|Polite=Form|PronType=Prs": {"morph": "Number=Plur|Person=2|Polite=Form|PronType=Prs", POS: PRON},
|
||||
"PRON__Number=Plur|Person=2|PronType=Prs": {"morph": "Number=Plur|Person=2|PronType=Prs", POS: PRON},
|
||||
"PRON__Number=Plur|Person=3|Poss=Yes|PronType=Prs": {"morph": "Number=Plur|Person=3|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Number=Plur|Person=3|PronType=Prs": {"morph": "Number=Plur|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Number=Plur|PronType=Dem": {"morph": "Number=Plur|PronType=Dem", POS: PRON},
|
||||
"PRON__Number=Plur|PronType=Ind": {"morph": "Number=Plur|PronType=Ind", POS: PRON},
|
||||
"PRON__Number=Plur|PronType=Int": {"morph": "Number=Plur|PronType=Int", POS: PRON},
|
||||
"PRON__Number=Plur|PronType=Rel": {"morph": "Number=Plur|PronType=Rel", POS: PRON},
|
||||
"PRON__Number=Sing|Person=1": {"morph": "Number=Sing|Person=1", POS: PRON},
|
||||
"PRON__Number=Sing|Person=1|PrepCase=Pre|PronType=Prs": {"morph": "Number=Sing|Person=1|PrepCase=Pre|PronType=Prs", POS: PRON},
|
||||
"PRON__Number=Sing|Person=1|PronType=Prs": {"morph": "Number=Sing|Person=1|PronType=Prs", POS: PRON},
|
||||
"PRON__Number=Sing|Person=2": {"morph": "Number=Sing|Person=2", POS: PRON},
|
||||
"PRON__Number=Sing|Person=2|Polite=Form|PronType=Prs": {"morph": "Number=Sing|Person=2|Polite=Form|PronType=Prs", POS: PRON},
|
||||
"PRON__Number=Sing|Person=2|PrepCase=Pre|PronType=Prs": {"morph": "Number=Sing|Person=2|PrepCase=Pre|PronType=Prs", POS: PRON},
|
||||
"PRON__Number=Sing|Person=2|PronType=Prs": {"morph": "Number=Sing|Person=2|PronType=Prs", POS: PRON},
|
||||
"PRON__Number=Sing|Person=3|Poss=Yes|PronType=Prs": {"morph": "Number=Sing|Person=3|Poss=Yes|PronType=Prs", POS: PRON},
|
||||
"PRON__Number=Sing|Person=3|PronType=Prs": {"morph": "Number=Sing|Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__Number=Sing|PronType=Dem": {"morph": "Number=Sing|PronType=Dem", POS: PRON},
|
||||
"PRON__Number=Sing|PronType=Ind": {"morph": "Number=Sing|PronType=Ind", POS: PRON},
|
||||
"PRON__Number=Sing|PronType=Int": {"morph": "Number=Sing|PronType=Int", POS: PRON},
|
||||
"PRON__Number=Sing|PronType=Rel": {"morph": "Number=Sing|PronType=Rel", POS: PRON},
|
||||
"PRON__Person=1|PronType=Prs": {"morph": "Person=1|PronType=Prs", POS: PRON},
|
||||
"PRON__Person=3": {"morph": "Person=3", POS: PRON},
|
||||
"PRON__Person=3|PrepCase=Pre|PronType=Prs": {"morph": "Person=3|PrepCase=Pre|PronType=Prs", POS: PRON},
|
||||
"PRON__Person=3|PronType=Prs": {"morph": "Person=3|PronType=Prs", POS: PRON},
|
||||
"PRON__PronType=Ind": {"morph": "PronType=Ind", POS: PRON},
|
||||
"PRON__PronType=Int": {"morph": "PronType=Int", POS: PRON},
|
||||
"PRON__PronType=Rel": {"morph": "PronType=Rel", POS: PRON},
|
||||
"PROPN___": {"morph": "_", POS: PROPN},
|
||||
"PUNCT___": {"morph": "_", POS: PUNCT},
|
||||
"PUNCT__PunctSide=Fin|PunctType=Brck": {"morph": "PunctSide=Fin|PunctType=Brck", POS: PUNCT},
|
||||
"PUNCT__PunctSide=Fin|PunctType=Excl": {"morph": "PunctSide=Fin|PunctType=Excl", POS: PUNCT},
|
||||
"PUNCT__PunctSide=Fin|PunctType=Qest": {"morph": "PunctSide=Fin|PunctType=Qest", POS: PUNCT},
|
||||
"PUNCT__PunctSide=Ini|PunctType=Brck": {"morph": "PunctSide=Ini|PunctType=Brck", POS: PUNCT},
|
||||
"PUNCT__PunctSide=Ini|PunctType=Excl": {"morph": "PunctSide=Ini|PunctType=Excl", POS: PUNCT},
|
||||
"PUNCT__PunctSide=Ini|PunctType=Qest": {"morph": "PunctSide=Ini|PunctType=Qest", POS: PUNCT},
|
||||
"PUNCT__PunctType=Colo": {"morph": "PunctType=Colo", POS: PUNCT},
|
||||
"PUNCT__PunctType=Comm": {"morph": "PunctType=Comm", POS: PUNCT},
|
||||
"PUNCT__PunctType=Dash": {"morph": "PunctType=Dash", POS: PUNCT},
|
||||
"PUNCT__PunctType=Peri": {"morph": "PunctType=Peri", POS: PUNCT},
|
||||
"PUNCT__PunctType=Quot": {"morph": "PunctType=Quot", POS: PUNCT},
|
||||
"PUNCT__PunctType=Semi": {"morph": "PunctType=Semi", POS: PUNCT},
|
||||
"SCONJ___": {"morph": "_", POS: SCONJ},
|
||||
"SYM___": {"morph": "_", POS: SYM},
|
||||
"SYM__NumForm=Digit": {"morph": "NumForm=Digit", POS: SYM},
|
||||
"SYM__NumForm=Digit|NumType=Frac": {"morph": "NumForm=Digit|NumType=Frac", POS: SYM},
|
||||
"VERB__Gender=Fem|Number=Plur|Tense=Past|VerbForm=Part": {"morph": "Gender=Fem|Number=Plur|Tense=Past|VerbForm=Part", POS: VERB},
|
||||
"VERB__Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part": {"morph": "Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part", POS: VERB},
|
||||
"VERB__Gender=Masc|Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {"morph": "Gender=Masc|Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Gender=Masc|Number=Plur|Tense=Past|VerbForm=Part": {"morph": "Gender=Masc|Number=Plur|Tense=Past|VerbForm=Part", POS: VERB},
|
||||
"VERB__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part": {"morph": "Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part", POS: VERB},
|
||||
"VERB__Mood=Cnd|Number=Plur|Person=1|VerbForm=Fin": {"morph": "Mood=Cnd|Number=Plur|Person=1|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Cnd|Number=Plur|Person=3|VerbForm=Fin": {"morph": "Mood=Cnd|Number=Plur|Person=3|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Cnd|Number=Sing|Person=1|VerbForm=Fin": {"morph": "Mood=Cnd|Number=Sing|Person=1|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Cnd|Number=Sing|Person=2|VerbForm=Fin": {"morph": "Mood=Cnd|Number=Sing|Person=2|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Cnd|Number=Sing|Person=3|VerbForm=Fin": {"morph": "Mood=Cnd|Number=Sing|Person=3|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Imp|Number=Plur|Person=1|VerbForm=Fin": {"morph": "Mood=Imp|Number=Plur|Person=1|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Imp|Number=Plur|Person=2|VerbForm=Fin": {"morph": "Mood=Imp|Number=Plur|Person=2|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Imp|Number=Plur|Person=3|VerbForm=Fin": {"morph": "Mood=Imp|Number=Plur|Person=3|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Imp|Number=Sing|Person=2|VerbForm=Fin": {"morph": "Mood=Imp|Number=Sing|Person=2|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Imp|Number=Sing|Person=3|VerbForm=Fin": {"morph": "Mood=Imp|Number=Sing|Person=3|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=1|Tense=Past|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=1|Tense=Past|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=3|Tense=Past|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=3|Tense=Past|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=1|Tense=Fut|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=1|Tense=Fut|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=1|Tense=Past|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=1|Tense=Past|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=2|Tense=Fut|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=2|Tense=Fut|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=2|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=2|Tense=Imp|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=2|Tense=Past|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=2|Tense=Past|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=3|Tense=Past|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=3|Tense=Past|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Person=3|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Ind|Person=3|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Sub|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Sub|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Sub|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Sub|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Sub|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Sub|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Sub|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Sub|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {"morph": "Mood=Sub|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {"morph": "Mood=Sub|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin", POS: VERB},
|
||||
"VERB__VerbForm=Ger": {"morph": "VerbForm=Ger", POS: VERB},
|
||||
"VERB__VerbForm=Inf": {"morph": "VerbForm=Inf", POS: VERB},
|
||||
"X___": {"morph": "_", POS: X},
|
||||
"___PunctType=Quot": {POS: PUNCT},
|
||||
"___VerbForm=Inf": {POS: VERB},
|
||||
"___Number=Sing|Person=2|PronType=Prs": {POS: PRON},
|
||||
"_SP": {"morph": "_", POS: SPACE},
|
||||
}
|
||||
# fmt: on
|
||||
|
|
@ -1,7 +1,6 @@
|
|||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .punctuation import TOKENIZER_SUFFIXES
|
||||
from .tag_map import TAG_MAP
|
||||
|
||||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
from ...language import Language
|
||||
|
|
@ -14,7 +13,6 @@ class BasqueDefaults(Language.Defaults):
|
|||
lex_attr_getters[LANG] = lambda text: "eu"
|
||||
|
||||
tokenizer_exceptions = BASE_EXCEPTIONS
|
||||
tag_map = TAG_MAP
|
||||
stop_words = STOP_WORDS
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
|
||||
|
|
|
|||
|
|
@ -1,68 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, SYM, ADJ, CCONJ, NUM, DET, ADV, ADP, X, VERB
|
||||
from ...symbols import NOUN, PROPN, PART, INTJ, SPACE, PRON
|
||||
|
||||
TAG_MAP = {
|
||||
".": {POS: PUNCT, "PunctType": "peri"},
|
||||
",": {POS: PUNCT, "PunctType": "comm"},
|
||||
"-LRB-": {POS: PUNCT, "PunctType": "brck", "PunctSide": "ini"},
|
||||
"-RRB-": {POS: PUNCT, "PunctType": "brck", "PunctSide": "fin"},
|
||||
"``": {POS: PUNCT, "PunctType": "quot", "PunctSide": "ini"},
|
||||
'""': {POS: PUNCT, "PunctType": "quot", "PunctSide": "fin"},
|
||||
"''": {POS: PUNCT, "PunctType": "quot", "PunctSide": "fin"},
|
||||
":": {POS: PUNCT},
|
||||
"$": {POS: SYM, "Other": {"SymType": "currency"}},
|
||||
"#": {POS: SYM, "Other": {"SymType": "numbersign"}},
|
||||
"AFX": {POS: ADJ, "Hyph": "yes"},
|
||||
"CC": {POS: CCONJ, "ConjType": "coor"},
|
||||
"CD": {POS: NUM, "NumType": "card"},
|
||||
"DT": {POS: DET},
|
||||
"EX": {POS: ADV, "AdvType": "ex"},
|
||||
"FW": {POS: X, "Foreign": "yes"},
|
||||
"HYPH": {POS: PUNCT, "PunctType": "dash"},
|
||||
"IN": {POS: ADP},
|
||||
"JJ": {POS: ADJ, "Degree": "pos"},
|
||||
"JJR": {POS: ADJ, "Degree": "comp"},
|
||||
"JJS": {POS: ADJ, "Degree": "sup"},
|
||||
"LS": {POS: PUNCT, "NumType": "ord"},
|
||||
"MD": {POS: VERB, "VerbType": "mod"},
|
||||
"NIL": {POS: ""},
|
||||
"NN": {POS: NOUN, "Number": "sing"},
|
||||
"NNP": {POS: PROPN, "NounType": "prop", "Number": "sing"},
|
||||
"NNPS": {POS: PROPN, "NounType": "prop", "Number": "plur"},
|
||||
"NNS": {POS: NOUN, "Number": "plur"},
|
||||
"PDT": {POS: ADJ, "AdjType": "pdt", "PronType": "prn"},
|
||||
"POS": {POS: PART, "Poss": "yes"},
|
||||
"PRP": {POS: PRON, "PronType": "prs"},
|
||||
"PRP$": {POS: ADJ, "PronType": "prs", "Poss": "yes"},
|
||||
"RB": {POS: ADV, "Degree": "pos"},
|
||||
"RBR": {POS: ADV, "Degree": "comp"},
|
||||
"RBS": {POS: ADV, "Degree": "sup"},
|
||||
"RP": {POS: PART},
|
||||
"SP": {POS: SPACE},
|
||||
"SYM": {POS: SYM},
|
||||
"TO": {POS: PART, "PartType": "inf", "VerbForm": "inf"},
|
||||
"UH": {POS: INTJ},
|
||||
"VB": {POS: VERB, "VerbForm": "inf"},
|
||||
"VBD": {POS: VERB, "VerbForm": "fin", "Tense": "past"},
|
||||
"VBG": {POS: VERB, "VerbForm": "part", "Tense": "pres", "Aspect": "prog"},
|
||||
"VBN": {POS: VERB, "VerbForm": "part", "Tense": "past", "Aspect": "perf"},
|
||||
"VBP": {POS: VERB, "VerbForm": "fin", "Tense": "pres"},
|
||||
"VBZ": {
|
||||
POS: VERB,
|
||||
"VerbForm": "fin",
|
||||
"Tense": "pres",
|
||||
"Number": "sing",
|
||||
"Person": 3,
|
||||
},
|
||||
"WDT": {POS: ADJ, "PronType": "int|rel"},
|
||||
"WP": {POS: NOUN, "PronType": "int|rel"},
|
||||
"WP$": {POS: ADJ, "Poss": "yes", "PronType": "int|rel"},
|
||||
"WRB": {POS: ADV, "PronType": "int|rel"},
|
||||
"ADD": {POS: X},
|
||||
"NFP": {POS: PUNCT},
|
||||
"GW": {POS: X},
|
||||
"XX": {POS: X},
|
||||
"BES": {POS: VERB},
|
||||
"HVS": {POS: VERB},
|
||||
"_SP": {POS: SPACE},
|
||||
}
|
||||
|
|
@ -5,7 +5,6 @@ from ..norm_exceptions import BASE_NORMS
|
|||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .tag_map import TAG_MAP
|
||||
from .punctuation import TOKENIZER_SUFFIXES
|
||||
from .syntax_iterators import SYNTAX_ITERATORS
|
||||
|
||||
|
|
@ -19,7 +18,6 @@ class PersianDefaults(Language.Defaults):
|
|||
lex_attr_getters[LANG] = lambda text: "fa"
|
||||
tokenizer_exceptions = update_exc(TOKENIZER_EXCEPTIONS)
|
||||
stop_words = STOP_WORDS
|
||||
tag_map = TAG_MAP
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
writing_system = {"direction": "rtl", "has_case": False, "has_letters": True}
|
||||
syntax_iterators = SYNTAX_ITERATORS
|
||||
|
|
|
|||
|
|
@ -1,36 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, ADJ, CONJ, NUM, DET, ADV, ADP, X, VERB
|
||||
from ...symbols import PRON, NOUN, PART, INTJ, AUX
|
||||
|
||||
|
||||
TAG_MAP = {
|
||||
"ADJ": {POS: ADJ},
|
||||
"ADJ_CMPR": {POS: ADJ},
|
||||
"ADJ_INO": {POS: ADJ},
|
||||
"ADJ_SUP": {POS: ADJ},
|
||||
"ADV": {POS: ADV},
|
||||
"ADV_COMP": {POS: ADV},
|
||||
"ADV_I": {POS: ADV},
|
||||
"ADV_LOC": {POS: ADV},
|
||||
"ADV_NEG": {POS: ADV},
|
||||
"ADV_TIME": {POS: ADV},
|
||||
"CLITIC": {POS: PART},
|
||||
"CON": {POS: CONJ},
|
||||
"CONJ": {POS: CONJ},
|
||||
"DELM": {POS: PUNCT},
|
||||
"DET": {POS: DET},
|
||||
"FW": {POS: X},
|
||||
"INT": {POS: INTJ},
|
||||
"N_PL": {POS: NOUN},
|
||||
"N_SING": {POS: NOUN},
|
||||
"N_VOC": {POS: NOUN},
|
||||
"NUM": {POS: NUM},
|
||||
"P": {POS: ADP},
|
||||
"PREV": {POS: ADP},
|
||||
"PRO": {POS: PRON},
|
||||
"V_AUX": {POS: AUX},
|
||||
"V_IMP": {POS: VERB},
|
||||
"V_PA": {POS: VERB},
|
||||
"V_PP": {POS: VERB},
|
||||
"V_PRS": {POS: VERB},
|
||||
"V_SUB": {POS: VERB},
|
||||
}
|
||||
|
|
@ -1,7 +1,6 @@
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS, TOKEN_MATCH
|
||||
from .punctuation import TOKENIZER_PREFIXES, TOKENIZER_INFIXES
|
||||
from .punctuation import TOKENIZER_SUFFIXES
|
||||
from .tag_map import TAG_MAP
|
||||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .lemmatizer import FrenchLemmatizer
|
||||
|
|
@ -23,7 +22,6 @@ class FrenchDefaults(Language.Defaults):
|
|||
Language.Defaults.lex_attr_getters[NORM], BASE_NORMS
|
||||
)
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
tag_map = TAG_MAP
|
||||
stop_words = STOP_WORDS
|
||||
prefixes = TOKENIZER_PREFIXES
|
||||
infixes = TOKENIZER_INFIXES
|
||||
|
|
|
|||
|
|
@ -1,216 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, ADJ, CCONJ, NUM, DET, ADV, ADP, X, VERB
|
||||
from ...symbols import NOUN, PROPN, PART, INTJ, SPACE, PRON, AUX, SCONJ
|
||||
|
||||
|
||||
TAG_MAP = {
|
||||
"ADJ__Gender=Fem|Number=Plur": {POS: ADJ},
|
||||
"ADJ__Gender=Fem|Number=Plur|NumType=Ord": {POS: ADJ},
|
||||
"ADJ__Gender=Fem|Number=Sing": {POS: ADJ},
|
||||
"ADJ__Gender=Fem|Number=Sing|NumType=Ord": {POS: ADJ},
|
||||
"ADJ__Gender=Masc": {POS: ADJ},
|
||||
"ADJ__Gender=Masc|Number=Plur": {POS: ADJ},
|
||||
"ADJ__Gender=Masc|Number=Plur|NumType=Ord": {POS: ADJ},
|
||||
"ADJ__Gender=Masc|Number=Sing": {POS: ADJ},
|
||||
"ADJ__Gender=Masc|Number=Sing|NumType=Card": {POS: ADJ},
|
||||
"ADJ__Gender=Masc|Number=Sing|NumType=Ord": {POS: ADJ},
|
||||
"ADJ__NumType=Card": {POS: ADJ},
|
||||
"ADJ__NumType=Ord": {POS: ADJ},
|
||||
"ADJ__Number=Plur": {POS: ADJ},
|
||||
"ADJ__Number=Sing": {POS: ADJ},
|
||||
"ADJ__Number=Sing|NumType=Ord": {POS: ADJ},
|
||||
"ADJ___": {POS: ADJ},
|
||||
"ADP__Gender=Fem|Number=Plur|Person=3": {POS: ADP},
|
||||
"ADP__Gender=Masc|Number=Plur|Person=3": {POS: ADP},
|
||||
"ADP__Gender=Masc|Number=Sing|Person=3": {POS: ADP},
|
||||
"ADP___": {POS: ADP},
|
||||
"ADV__Polarity=Neg": {POS: ADV},
|
||||
"ADV__PronType=Int": {POS: ADV},
|
||||
"ADV___": {POS: ADV},
|
||||
"AUX__Gender=Fem|Number=Plur|Tense=Past|VerbForm=Part": {POS: AUX},
|
||||
"AUX__Gender=Fem|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass": {POS: AUX},
|
||||
"AUX__Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part": {POS: AUX},
|
||||
"AUX__Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass": {POS: AUX},
|
||||
"AUX__Gender=Masc|Number=Plur|Tense=Past|VerbForm=Part": {POS: AUX},
|
||||
"AUX__Gender=Masc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass": {POS: AUX},
|
||||
"AUX__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part": {POS: AUX},
|
||||
"AUX__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass": {POS: AUX},
|
||||
"AUX__Mood=Cnd|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Cnd|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Cnd|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Cnd|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Cnd|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Imp|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=3|Tense=Past|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=2|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=3|Tense=Past|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Mood=Sub|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"AUX__Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass": {POS: AUX},
|
||||
"AUX__Tense=Past|VerbForm=Part": {POS: AUX},
|
||||
"AUX__Tense=Pres|VerbForm=Part": {POS: AUX},
|
||||
"AUX__VerbForm=Inf": {POS: AUX},
|
||||
"CCONJ___": {POS: CCONJ},
|
||||
"DET__Definite=Def|Gender=Fem|Number=Sing|PronType=Art": {POS: DET},
|
||||
"DET__Definite=Def|Gender=Masc|Number=Sing|PronType=Art": {POS: DET},
|
||||
"DET__Definite=Def|Number=Plur|PronType=Art": {POS: DET},
|
||||
"DET__Definite=Def|Number=Sing|PronType=Art": {POS: DET},
|
||||
"DET__Definite=Ind|Gender=Fem|Number=Plur|PronType=Art": {POS: DET},
|
||||
"DET__Definite=Ind|Gender=Fem|Number=Sing|PronType=Art": {POS: DET},
|
||||
"DET__Definite=Ind|Gender=Masc|Number=Plur|PronType=Art": {POS: DET},
|
||||
"DET__Definite=Ind|Gender=Masc|Number=Sing|PronType=Art": {POS: DET},
|
||||
"DET__Definite=Ind|Number=Plur|PronType=Art": {POS: DET},
|
||||
"DET__Definite=Ind|Number=Sing|PronType=Art": {POS: DET},
|
||||
"DET__Gender=Fem|Number=Plur": {POS: DET},
|
||||
"DET__Gender=Fem|Number=Plur|PronType=Int": {POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing": {POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing|Poss=Yes": {POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Dem": {POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Int": {POS: DET},
|
||||
"DET__Gender=Masc|Number=Plur": {POS: DET},
|
||||
"DET__Gender=Masc|Number=Sing": {POS: DET},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Dem": {POS: DET},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Int": {POS: DET},
|
||||
"DET__Number=Plur": {POS: DET},
|
||||
"DET__Number=Plur|Poss=Yes": {POS: DET},
|
||||
"DET__Number=Plur|PronType=Dem": {POS: DET},
|
||||
"DET__Number=Sing": {POS: DET},
|
||||
"DET__Number=Sing|Poss=Yes": {POS: DET},
|
||||
"DET___": {POS: DET},
|
||||
"INTJ___": {POS: INTJ},
|
||||
"NOUN__Gender=Fem": {POS: NOUN},
|
||||
"NOUN__Gender=Fem|Number=Plur": {POS: NOUN},
|
||||
"NOUN__Gender=Fem|Number=Sing": {POS: NOUN},
|
||||
"NOUN__Gender=Masc": {POS: NOUN},
|
||||
"NOUN__Gender=Masc|Number=Plur": {POS: NOUN},
|
||||
"NOUN__Gender=Masc|Number=Plur|NumType=Card": {POS: NOUN},
|
||||
"NOUN__Gender=Masc|Number=Sing": {POS: NOUN},
|
||||
"NOUN__Gender=Masc|Number=Sing|NumType=Card": {POS: NOUN},
|
||||
"NOUN__NumType=Card": {POS: NOUN},
|
||||
"NOUN__Number=Plur": {POS: NOUN},
|
||||
"NOUN__Number=Sing": {POS: NOUN},
|
||||
"NOUN___": {POS: NOUN},
|
||||
"NUM__Gender=Masc|Number=Plur|NumType=Card": {POS: NUM},
|
||||
"NUM__NumType=Card": {POS: NUM},
|
||||
"PART___": {POS: PART},
|
||||
"PRON__Gender=Fem|Number=Plur": {POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Plur|Person=3": {POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Plur|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Plur|Person=3|PronType=Rel": {POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Plur|PronType=Dem": {POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Plur|PronType=Rel": {POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|Person=3": {POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|PronType=Dem": {POS: PRON},
|
||||
"PRON__Gender=Fem|Number=Sing|PronType=Rel": {POS: PRON},
|
||||
"PRON__Gender=Fem|PronType=Rel": {POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur": {POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur|Person=3": {POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur|Person=3|PronType=Rel": {POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur|PronType=Dem": {POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Plur|PronType=Rel": {POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing": {POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|Person=3": {POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|Person=3|PronType=Dem": {POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|PronType=Dem": {POS: PRON},
|
||||
"PRON__Gender=Masc|Number=Sing|PronType=Rel": {POS: PRON},
|
||||
"PRON__Gender=Masc|PronType=Rel": {POS: PRON},
|
||||
"PRON__NumType=Card|PronType=Rel": {POS: PRON},
|
||||
"PRON__Number=Plur|Person=1": {POS: PRON},
|
||||
"PRON__Number=Plur|Person=1|PronType=Prs": {POS: PRON},
|
||||
"PRON__Number=Plur|Person=1|Reflex=Yes": {POS: PRON},
|
||||
"PRON__Number=Plur|Person=2": {POS: PRON},
|
||||
"PRON__Number=Plur|Person=2|PronType=Prs": {POS: PRON},
|
||||
"PRON__Number=Plur|Person=2|Reflex=Yes": {POS: PRON},
|
||||
"PRON__Number=Plur|Person=3": {POS: PRON},
|
||||
"PRON__Number=Plur|PronType=Rel": {POS: PRON},
|
||||
"PRON__Number=Sing|Person=1": {POS: PRON},
|
||||
"PRON__Number=Sing|Person=1|PronType=Prs": {POS: PRON},
|
||||
"PRON__Number=Sing|Person=1|Reflex=Yes": {POS: PRON},
|
||||
"PRON__Number=Sing|Person=2|PronType=Prs": {POS: PRON},
|
||||
"PRON__Number=Sing|Person=3": {POS: PRON},
|
||||
"PRON__Number=Sing|PronType=Dem": {POS: PRON},
|
||||
"PRON__Number=Sing|PronType=Rel": {POS: PRON},
|
||||
"PRON__Person=3": {POS: PRON},
|
||||
"PRON__Person=3|Reflex=Yes": {POS: PRON},
|
||||
"PRON__PronType=Int": {POS: PRON},
|
||||
"PRON__PronType=Rel": {POS: PRON},
|
||||
"PRON___": {POS: PRON},
|
||||
"PROPN__Gender=Fem|Number=Plur": {POS: PROPN},
|
||||
"PROPN__Gender=Fem|Number=Sing": {POS: PROPN},
|
||||
"PROPN__Gender=Masc": {POS: PROPN},
|
||||
"PROPN__Gender=Masc|Number=Plur": {POS: PROPN},
|
||||
"PROPN__Gender=Masc|Number=Sing": {POS: PROPN},
|
||||
"PROPN__Number=Plur": {POS: PROPN},
|
||||
"PROPN__Number=Sing": {POS: PROPN},
|
||||
"PROPN___": {POS: PROPN},
|
||||
"PUNCT___": {POS: PUNCT},
|
||||
"SCONJ___": {POS: SCONJ},
|
||||
"VERB__Gender=Fem|Number=Plur|Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"VERB__Gender=Fem|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass": {POS: VERB},
|
||||
"VERB__Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"VERB__Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass": {POS: VERB},
|
||||
"VERB__Gender=Masc|Number=Plur|Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"VERB__Gender=Masc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass": {POS: VERB},
|
||||
"VERB__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"VERB__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass": {POS: VERB},
|
||||
"VERB__Gender=Masc|Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"VERB__Gender=Masc|Tense=Past|VerbForm=Part|Voice=Pass": {POS: VERB},
|
||||
"VERB__Mood=Cnd|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Cnd|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Cnd|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Imp|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Imp|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=2|Tense=Fut|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=2|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=3|Tense=Past|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=1|Tense=Fut|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=3|Tense=Past|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|Person=3|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Ind|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Sing|Person=3|Tense=Past|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Mood=Sub|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"VERB__Number=Plur|Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"VERB__Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass": {POS: VERB},
|
||||
"VERB__Number=Sing|Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"VERB__Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass": {POS: VERB},
|
||||
"VERB__Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"VERB__Tense=Past|VerbForm=Part|Voice=Pass": {POS: VERB},
|
||||
"VERB__Tense=Pres|VerbForm=Part": {POS: VERB},
|
||||
"VERB__VerbForm=Inf": {POS: VERB},
|
||||
"VERB__VerbForm=Part": {POS: VERB},
|
||||
"X___": {POS: X},
|
||||
"_SP": {POS: SPACE},
|
||||
}
|
||||
|
|
@ -1,6 +1,5 @@
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .stop_words import STOP_WORDS
|
||||
from .tag_map import TAG_MAP
|
||||
|
||||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
from ...language import Language
|
||||
|
|
@ -14,7 +13,6 @@ class IrishDefaults(Language.Defaults):
|
|||
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
stop_words = set(STOP_WORDS)
|
||||
tag_map = TAG_MAP
|
||||
|
||||
|
||||
class Irish(Language):
|
||||
|
|
|
|||
|
|
@ -1,366 +0,0 @@
|
|||
# fmt: off
|
||||
TAG_MAP = {
|
||||
"ADJ__Case=Gen|Form=Len|Gender=Masc|Number=Sing": {"pos": "ADJ", "Case": "gen", "Gender": "masc", "Number": "sing", "Form": "len"},
|
||||
"ADJ__Case=Gen|Gender=Fem|Number=Sing": {"pos": "ADJ", "Case": "gen", "Gender": "fem", "Number": "sing"},
|
||||
"ADJ__Case=Gen|Gender=Masc|Number=Sing": {"pos": "ADJ", "Case": "gen", "Gender": "masc", "Number": "sing"},
|
||||
"ADJ__Case=Gen|NounType=Strong|Number=Plur": {"pos": "ADJ", "Case": "gen", "Number": "plur", "NounType": "strong"},
|
||||
"ADJ__Case=Gen|NounType=Weak|Number=Plur": {"pos": "ADJ", "Case": "gen", "Number": "plur", "NounType": "weak"},
|
||||
"ADJ__Case=NomAcc|Form=Len|Gender=Fem|Number=Sing": {"pos": "ADJ", "Case": "nom|acc", "Gender": "fem", "Number": "sing", "Form": "len"},
|
||||
"ADJ__Case=NomAcc|Form=Len|Gender=Masc|Number=Sing": {"pos": "ADJ", "Case": "nom|acc", "Gender": "masc", "Number": "sing", "Form": "len"},
|
||||
"ADJ__Case=NomAcc|Gender=Fem|Number=Plur": {"pos": "ADJ", "Case": "nom|acc", "Gender": "fem", "Number": "plur"},
|
||||
"ADJ__Case=NomAcc|Gender=Fem|Number=Sing": {"pos": "ADJ", "Case": "nom|acc", "Gender": "fem", "Number": "sing"},
|
||||
"ADJ__Case=NomAcc|Gender=Masc|Number=Plur": {"pos": "ADJ", "Case": "nom|acc", "Gender": "masc", "Number": "plur"},
|
||||
"ADJ__Case=NomAcc|Gender=Masc|Number=Sing": {"pos": "ADJ", "Case": "nom|acc", "Gender": "masc", "Number": "sing"},
|
||||
"ADJ__Case=NomAcc|NounType=NotSlender|Number=Plur": {"pos": "ADJ", "Case": "nom|acc", "Number": "plur", "NounType": "notslender"},
|
||||
"ADJ__Case=NomAcc|NounType=Slender|Number=Plur": {"pos": "ADJ", "Case": "nom|acc", "Number": "plur", "NounType": "slender"},
|
||||
"ADJ__Degree=Cmp,Sup|Form=Len": {"pos": "ADJ", "Degree": "cmp|sup", "Form": "len"},
|
||||
"ADJ__Degree=Cmp,Sup": {"pos": "ADJ", "Degree": "cmp|sup"},
|
||||
"ADJ__Degree=Pos|Form=Ecl": {"pos": "ADJ", "Degree": "pos", "Form": "ecl"},
|
||||
"ADJ__Degree=Pos|Form=HPref": {"pos": "ADJ", "Degree": "pos", "Form": "hpref"},
|
||||
"ADJ__Degree=Pos|Form=Len": {"pos": "ADJ", "Degree": "pos", "Form": "len"},
|
||||
"ADJ__Degree=Pos": {"pos": "ADJ", "Degree": "pos"},
|
||||
"ADJ__Foreign=Yes": {"pos": "ADJ", "Foreign": "yes"},
|
||||
"ADJ__Form=Len|VerbForm=Part": {"pos": "ADJ", "VerbForm": "part", "Form": "len"},
|
||||
"ADJ__Gender=Masc|Number=Sing|PartType=Voc": {"pos": "ADJ", "Gender": "masc", "Number": "sing", "Case": "voc"},
|
||||
"ADJ__Gender=Masc|Number=Sing|Case=Voc": {"pos": "ADJ", "Gender": "masc", "Number": "sing", "Case": "voc"},
|
||||
"ADJ__Number=Plur|PartType=Voc": {"pos": "ADJ", "Number": "plur", "Case": "voc"},
|
||||
"ADJ__Number=Plur|Case=Voc": {"pos": "ADJ", "Number": "plur", "Case": "voc"},
|
||||
"ADJ__Number=Plur": {"pos": "ADJ", "Number": "plur"},
|
||||
"ADJ___": {"pos": "ADJ"},
|
||||
"ADJ__VerbForm=Part": {"pos": "ADJ", "VerbForm": "part"},
|
||||
"ADP__Foreign=Yes": {"pos": "ADP", "Foreign": "yes"},
|
||||
"ADP__Form=Len|Number=Plur|Person=1": {"pos": "ADP", "Number": "plur", "Person": 1, "Form": "len"},
|
||||
"ADP__Form=Len|Number=Plur|Person=3": {"pos": "ADP", "Number": "plur", "Person": 3, "Form": "len"},
|
||||
"ADP__Form=Len|Number=Sing|Person=1": {"pos": "ADP", "Number": "sing", "Person": 1, "Form": "len"},
|
||||
"ADP__Gender=Fem|Number=Sing|Person=3": {"pos": "ADP", "Gender": "fem", "Number": "sing", "Person": 3},
|
||||
"ADP__Gender=Fem|Number=Sing|Person=3|Poss=Yes": {"pos": "ADP", "Gender": "fem", "Number": "sing", "Person": 3, "Poss": "yes"},
|
||||
"ADP__Gender=Fem|Number=Sing|Person=3|Poss=Yes|PronType=Prs": {"pos": "ADP", "Gender": "fem", "Number": "sing", "Person": 3, "Poss": "yes", "PronType": "prs"},
|
||||
"ADP__Gender=Masc|Number=Sing|Person=3": {"pos": "ADP", "Gender": "masc", "Number": "sing", "Person": 3},
|
||||
"ADP__Gender=Masc|Number=Sing|Person=3|Poss=Yes": {"pos": "ADP", "Gender": "masc", "Number": "sing", "Person": 3, "Poss": "yes"},
|
||||
"ADP__Gender=Masc|Number=Sing|Person=3|Poss=Yes|PronType=Prs": {"pos": "ADP", "Gender": "masc", "Number": "sing", "Person": 3, "Poss": "yes", "PronType": "prs"},
|
||||
"ADP__Gender=Masc|Number=Sing|Person=3|PronType=Emp": {"pos": "ADP", "Gender": "masc", "Number": "sing", "Person": 3, "PronType": "emp"},
|
||||
"ADP__Number=Plur|Person=1": {"pos": "ADP", "Number": "plur", "Person": 1},
|
||||
"ADP__Number=Plur|Person=1|Poss=Yes": {"pos": "ADP", "Number": "plur", "Person": 1, "Poss": "yes"},
|
||||
"ADP__Number=Plur|Person=1|PronType=Emp": {"pos": "ADP", "Number": "plur", "Person": 1, "PronType": "emp"},
|
||||
"ADP__Number=Plur|Person=2": {"pos": "ADP", "Number": "plur", "Person": 2},
|
||||
"ADP__Number=Plur|Person=3": {"pos": "ADP", "Number": "plur", "Person": 3},
|
||||
"ADP__Number=Plur|Person=3|Poss=Yes": {"pos": "ADP", "Number": "plur", "Person": 3, "Poss": "yes"},
|
||||
"ADP__Number=Plur|Person=3|Poss=Yes|PronType=Prs": {"pos": "ADP", "Number": "plur", "Person": 3, "Poss": "yes", "PronType": "prs"},
|
||||
"ADP__Number=Plur|Person=3|PronType=Emp": {"pos": "ADP", "Number": "plur", "Person": 3, "PronType": "emp"},
|
||||
"ADP__Number=Plur|PronType=Art": {"pos": "ADP", "Number": "plur", "PronType": "art"},
|
||||
"ADP__Number=Sing|Person=1": {"pos": "ADP", "Number": "sing", "Person": 1},
|
||||
"ADP__Number=Sing|Person=1|Poss=Yes": {"pos": "ADP", "Number": "sing", "Person": 1, "Poss": "yes"},
|
||||
"ADP__Number=Sing|Person=1|PronType=Emp": {"pos": "ADP", "Number": "sing", "Person": 1, "PronType": "emp"},
|
||||
"ADP__Number=Sing|Person=2": {"pos": "ADP", "Number": "sing", "Person": 2},
|
||||
"ADP__Number=Sing|Person=3": {"pos": "ADP", "Number": "sing", "Person": 3},
|
||||
"ADP__Number=Sing|PronType=Art": {"pos": "ADP", "Number": "sing", "PronType": "art"},
|
||||
"ADP__Person=3|Poss=Yes": {"pos": "ADP", "Person": 3, "Poss": "yes"},
|
||||
"ADP___": {"pos": "ADP"},
|
||||
"ADP__Poss=Yes": {"pos": "ADP", "Poss": "yes"},
|
||||
"ADP__PrepForm=Cmpd": {"pos": "ADP", "PrepForm": "cmpd"},
|
||||
"ADP__PronType=Art": {"pos": "ADP", "PronType": "art"},
|
||||
"ADV__Form=Len": {"pos": "ADV", "Form": "len"},
|
||||
"ADV___": {"pos": "ADV"},
|
||||
"ADV__PronType=Int": {"pos": "ADV", "PronType": "int"},
|
||||
"AUX__Form=VF|Polarity=Neg|PronType=Rel|Tense=Past|VerbForm=Cop": {"pos": "AUX", "Polarity": "neg", "PronType": "rel", "Tense": "past", "Form": "vf", "VerbForm": "cop"},
|
||||
"AUX__Form=VF|Polarity=Neg|Tense=Past|VerbForm=Cop": {"pos": "AUX", "Polarity": "neg", "Tense": "past", "Form": "vf", "VerbForm": "cop"},
|
||||
"AUX__Form=VF|PronType=Rel|Tense=Past|VerbForm=Cop": {"pos": "AUX", "PronType": "rel", "Tense": "past", "Form": "vf", "VerbForm": "cop"},
|
||||
"AUX__Form=VF|Tense=Past|VerbForm=Cop": {"pos": "AUX", "Tense": "past", "Form": "vf", "VerbForm": "cop"},
|
||||
"AUX__Form=VF|Tense=Pres|VerbForm=Cop": {"pos": "AUX", "Tense": "pres", "Form": "vf", "VerbForm": "cop"},
|
||||
"AUX__Gender=Masc|Number=Sing|Person=3|VerbForm=Cop": {"pos": "AUX", "Gender": "masc", "Number": "sing", "Person": 3, "VerbForm": "cop"},
|
||||
"AUX__Mood=Int|Number=Sing|PronType=Art|VerbForm=Cop": {"pos": "AUX", "Number": "sing", "PronType": "art", "Mood": "int", "VerbForm": "cop"},
|
||||
"AUX__Mood=Int|Polarity=Neg|Tense=Past|VerbForm=Cop": {"pos": "AUX", "Polarity": "neg", "Tense": "past", "Mood": "int", "VerbForm": "cop"},
|
||||
"AUX__Mood=Int|Polarity=Neg|Tense=Pres|VerbForm=Cop": {"pos": "AUX", "Polarity": "neg", "Tense": "pres", "Mood": "int", "VerbForm": "cop"},
|
||||
"AUX__Mood=Int|Tense=Pres|VerbForm=Cop": {"pos": "AUX", "Tense": "pres", "Mood": "int", "VerbForm": "cop"},
|
||||
"AUX__PartType=Comp|Tense=Past|VerbForm=Cop": {"pos": "AUX", "Tense": "past", "PartType": "comp", "VerbForm": "cop"},
|
||||
"AUX__Polarity=Neg|PronType=Rel|Tense=Past|VerbForm=Cop": {"pos": "AUX", "Polarity": "neg", "PronType": "rel", "Tense": "past", "VerbForm": "cop"},
|
||||
"AUX__Polarity=Neg|PronType=Rel|Tense=Pres|VerbForm=Cop": {"pos": "AUX", "Polarity": "neg", "PronType": "rel", "Tense": "pres", "VerbForm": "cop"},
|
||||
"AUX__Polarity=Neg|Tense=Past|VerbForm=Cop": {"pos": "AUX", "Polarity": "neg", "Tense": "past", "VerbForm": "cop"},
|
||||
"AUX__Polarity=Neg|Tense=Pres|VerbForm=Cop": {"pos": "AUX", "Polarity": "neg", "Tense": "pres", "VerbForm": "cop"},
|
||||
"AUX___": {"pos": "AUX"},
|
||||
"AUX__PronType=Dem|VerbForm=Cop": {"pos": "AUX", "PronType": "dem", "VerbForm": "cop"},
|
||||
"AUX__PronType=Rel|Tense=Past|VerbForm=Cop": {"pos": "AUX", "PronType": "rel", "Tense": "past", "VerbForm": "cop"},
|
||||
"AUX__PronType=Rel|Tense=Pres|VerbForm=Cop": {"pos": "AUX", "PronType": "rel", "Tense": "pres", "VerbForm": "cop"},
|
||||
"AUX__Tense=Past|VerbForm=Cop": {"pos": "AUX", "Tense": "past", "VerbForm": "cop"},
|
||||
"AUX__Tense=Pres|VerbForm=Cop": {"pos": "AUX", "Tense": "pres", "VerbForm": "cop"},
|
||||
"AUX__VerbForm=Cop": {"pos": "AUX", "VerbForm": "cop"},
|
||||
"CCONJ___": {"pos": "CCONJ"},
|
||||
"DET__Case=Gen|Definite=Def|Gender=Fem|Number=Sing|PronType=Art": {"pos": "DET", "Case": "gen", "Definite": "def", "Gender": "fem", "Number": "sing", "PronType": "art"},
|
||||
"DET__Definite=Def|Form=Ecl": {"pos": "DET", "Definite": "def", "Form": "ecl"},
|
||||
"DET__Definite=Def|Gender=Fem|Number=Sing|PronType=Art": {"pos": "DET", "Definite": "def", "Gender": "fem", "Number": "sing", "PronType": "art"},
|
||||
"DET__Definite=Def|Number=Plur|PronType=Art": {"pos": "DET", "Definite": "def", "Number": "plur", "PronType": "art"},
|
||||
"DET__Definite=Def|Number=Sing|PronType=Art": {"pos": "DET", "Definite": "def", "Number": "sing", "PronType": "art"},
|
||||
"DET__Definite=Def": {"pos": "DET", "Definite": "def"},
|
||||
"DET__Form=HPref|PronType=Ind": {"pos": "DET", "PronType": "ind", "Form": "hpref"},
|
||||
"DET__Gender=Fem|Number=Sing|Person=3|Poss=Yes": {"pos": "DET", "Gender": "fem", "Number": "sing", "Person": 3, "Poss": "yes"},
|
||||
"DET__Gender=Masc|Number=Sing|Person=3|Poss=Yes": {"pos": "DET", "Gender": "masc", "Number": "sing", "Person": 3, "Poss": "yes"},
|
||||
"DET__Number=Plur|Person=1|Poss=Yes": {"pos": "DET", "Number": "plur", "Person": 1, "Poss": "yes"},
|
||||
"DET__Number=Plur|Person=3|Poss=Yes": {"pos": "DET", "Number": "plur", "Person": 3, "Poss": "yes"},
|
||||
"DET__Number=Sing|Person=1|Poss=Yes": {"pos": "DET", "Number": "sing", "Person": 1, "Poss": "yes"},
|
||||
"DET__Number=Sing|Person=2|Poss=Yes": {"pos": "DET", "Number": "sing", "Person": 2, "Poss": "yes"},
|
||||
"DET__Number=Sing|PronType=Int": {"pos": "DET", "Number": "sing", "PronType": "int"},
|
||||
"DET___": {"pos": "DET"},
|
||||
"DET__PronType=Dem": {"pos": "DET", "PronType": "dem"},
|
||||
"DET__PronType=Ind": {"pos": "DET", "PronType": "ind"},
|
||||
"NOUN__Case=Dat|Definite=Ind|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "dat", "Definite": "ind", "Gender": "fem", "Number": "sing"},
|
||||
"NOUN__Case=Dat|Form=Ecl|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "dat", "Gender": "fem", "Number": "sing", "Form": "ecl"},
|
||||
"NOUN__Case=Dat|Form=Len|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "dat", "Gender": "fem", "Number": "sing", "Form": "len"},
|
||||
"NOUN__Case=Dat|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "dat", "Gender": "fem", "Number": "sing"},
|
||||
"NOUN__Case=Dat|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "dat", "Gender": "masc", "Number": "sing"},
|
||||
"NOUN__Case=Gen|Definite=Def|Gender=Fem|NounType=Strong|Number=Plur": {"pos": "NOUN", "Case": "gen", "Definite": "def", "Gender": "fem", "Number": "plur", "NounType": "strong"},
|
||||
"NOUN__Case=Gen|Definite=Def|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "gen", "Definite": "def", "Gender": "fem", "Number": "sing"},
|
||||
"NOUN__Case=Gen|Definite=Def|Gender=Masc|NounType=Strong|Number=Plur": {"pos": "NOUN", "Case": "gen", "Definite": "def", "Gender": "masc", "Number": "plur", "NounType": "strong"},
|
||||
"NOUN__Case=Gen|Definite=Def|Gender=Masc|NounType=Weak|Number=Plur": {"pos": "NOUN", "Case": "gen", "Definite": "def", "Gender": "masc", "Number": "plur", "NounType": "weak"},
|
||||
"NOUN__Case=Gen|Definite=Def|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "gen", "Definite": "def", "Gender": "masc", "Number": "sing"},
|
||||
"NOUN__Case=Gen|Definite=Ind|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "gen", "Definite": "ind", "Gender": "fem", "Number": "sing"},
|
||||
"NOUN__Case=Gen|Form=Ecl|Gender=Fem|NounType=Strong|Number=Plur": {"pos": "NOUN", "Case": "gen", "Gender": "fem", "Number": "plur", "Form": "ecl", "NounType": "strong"},
|
||||
"NOUN__Case=Gen|Form=Ecl|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "gen", "Gender": "fem", "Number": "sing", "Form": "ecl"},
|
||||
"NOUN__Case=Gen|Form=Ecl|Gender=Masc|NounType=Strong|Number=Plur": {"pos": "NOUN", "Case": "gen", "Gender": "masc", "Number": "plur", "Form": "ecl", "NounType": "strong"},
|
||||
"NOUN__Case=Gen|Form=Ecl|Gender=Masc|NounType=Weak|Number=Plur": {"pos": "NOUN", "Case": "gen", "Gender": "masc", "Number": "plur", "Form": "ecl", "NounType": "weak"},
|
||||
"NOUN__Case=Gen|Form=Ecl|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "gen", "Gender": "masc", "Number": "sing", "Form": "ecl"},
|
||||
"NOUN__Case=Gen|Form=HPref|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "gen", "Gender": "fem", "Number": "sing", "Form": "hpref"},
|
||||
"NOUN__Case=Gen|Form=Len|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "gen", "Gender": "fem", "Number": "sing", "Form": "len"},
|
||||
"NOUN__Case=Gen|Form=Len|Gender=Masc|NounType=Strong|Number=Plur": {"pos": "NOUN", "Case": "gen", "Gender": "masc", "Number": "plur", "Form": "len", "NounType": "strong"},
|
||||
"NOUN__Case=Gen|Form=Len|Gender=Masc|NounType=Weak|Number=Plur": {"pos": "NOUN", "Case": "gen", "Gender": "masc", "Number": "plur", "Form": "len", "NounType": "weak"},
|
||||
"NOUN__Case=Gen|Form=Len|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "gen", "Gender": "masc", "Number": "sing", "Form": "len"},
|
||||
"NOUN__Case=Gen|Form=Len|VerbForm=Inf": {"pos": "NOUN", "Case": "gen", "VerbForm": "inf", "Form": "len"},
|
||||
"NOUN__Case=Gen|Gender=Fem|NounType=Strong|Number=Plur": {"pos": "NOUN", "Case": "gen", "Gender": "fem", "Number": "plur", "NounType": "strong"},
|
||||
"NOUN__Case=Gen|Gender=Fem|NounType=Weak|Number=Plur": {"pos": "NOUN", "Case": "gen", "Gender": "fem", "Number": "plur", "NounType": "weak"},
|
||||
"NOUN__Case=Gen|Gender=Fem|Number=Plur": {"pos": "NOUN", "Case": "gen", "Gender": "fem", "Number": "plur"},
|
||||
"NOUN__Case=Gen|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "gen", "Gender": "fem", "Number": "sing"},
|
||||
"NOUN__Case=Gen|Gender=Masc|NounType=Strong|Number=Plur": {"pos": "NOUN", "Case": "gen", "Gender": "masc", "Number": "plur", "NounType": "strong"},
|
||||
"NOUN__Case=Gen|Gender=Masc|NounType=Weak|Number=Plur": {"pos": "NOUN", "Case": "gen", "Gender": "masc", "Number": "plur", "NounType": "weak"},
|
||||
"NOUN__Case=Gen|Gender=Masc|Number=Plur": {"pos": "NOUN", "Case": "gen", "Gender": "masc", "Number": "plur"},
|
||||
"NOUN__Case=Gen|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "gen", "Gender": "masc", "Number": "sing"},
|
||||
"NOUN__Case=Gen|Number=Sing": {"pos": "NOUN", "Case": "gen", "Number": "sing"},
|
||||
"NOUN__Case=Gen|VerbForm=Inf": {"pos": "NOUN", "Case": "gen", "VerbForm": "inf"},
|
||||
"NOUN__Case=NomAcc|Definite=Def|Gender=Fem|Number=Plur": {"pos": "NOUN", "Case": "nom|acc", "Definite": "def", "Gender": "fem", "Number": "plur"},
|
||||
"NOUN__Case=NomAcc|Definite=Def|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "nom|acc", "Definite": "def", "Gender": "fem", "Number": "sing"},
|
||||
"NOUN__Case=NomAcc|Definite=Def|Gender=Fem": {"pos": "NOUN", "Case": "nom|acc", "Definite": "def", "Gender": "fem"},
|
||||
"NOUN__Case=NomAcc|Definite=Def|Gender=Masc|Number=Plur": {"pos": "NOUN", "Case": "nom|acc", "Definite": "def", "Gender": "masc", "Number": "plur"},
|
||||
"NOUN__Case=NomAcc|Definite=Def|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "nom|acc", "Definite": "def", "Gender": "masc", "Number": "sing"},
|
||||
"NOUN__Case=NomAcc|Definite=Ind|Gender=Masc|Number=Plur": {"pos": "NOUN", "Case": "nom|acc", "Definite": "ind", "Gender": "masc", "Number": "plur"},
|
||||
"NOUN__Case=NomAcc|Form=Ecl|Gender=Fem|Number=Plur": {"pos": "NOUN", "Case": "nom|acc", "Gender": "fem", "Number": "plur", "Form": "ecl"},
|
||||
"NOUN__Case=NomAcc|Form=Ecl|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "nom|acc", "Gender": "fem", "Number": "sing", "Form": "ecl"},
|
||||
"NOUN__Case=NomAcc|Form=Ecl|Gender=Masc|Number=Plur": {"pos": "NOUN", "Case": "nom|acc", "Gender": "masc", "Number": "plur", "Form": "ecl"},
|
||||
"NOUN__Case=NomAcc|Form=Ecl|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "nom|acc", "Gender": "masc", "Number": "sing", "Form": "ecl"},
|
||||
"NOUN__Case=NomAcc|Form=Emp|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "nom|acc", "Gender": "masc", "Number": "sing", "Form": "emp"},
|
||||
"NOUN__Case=NomAcc|Form=HPref|Gender=Fem|Number=Plur": {"pos": "NOUN", "Case": "nom|acc", "Gender": "fem", "Number": "plur", "Form": "hpref"},
|
||||
"NOUN__Case=NomAcc|Form=HPref|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "nom|acc", "Gender": "fem", "Number": "sing", "Form": "hpref"},
|
||||
"NOUN__Case=NomAcc|Form=HPref|Gender=Masc|Number=Plur": {"pos": "NOUN", "Case": "nom|acc", "Gender": "masc", "Number": "plur", "Form": "hpref"},
|
||||
"NOUN__Case=NomAcc|Form=HPref|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "nom|acc", "Gender": "masc", "Number": "sing", "Form": "hpref"},
|
||||
"NOUN__Case=NomAcc|Form=Len|Gender=Fem|Number=Plur": {"pos": "NOUN", "Case": "nom|acc", "Gender": "fem", "Number": "plur", "Form": "len"},
|
||||
"NOUN__Case=NomAcc|Form=Len|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "nom|acc", "Gender": "fem", "Number": "sing", "Form": "len"},
|
||||
"NOUN__Case=NomAcc|Form=Len|Gender=Masc|Number=Plur": {"pos": "NOUN", "Case": "nom|acc", "Gender": "masc", "Number": "plur", "Form": "len"},
|
||||
"NOUN__Case=NomAcc|Form=Len|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "nom|acc", "Gender": "masc", "Number": "sing", "Form": "len"},
|
||||
"NOUN__Case=NomAcc|Gender=Fem|Number=Plur": {"pos": "NOUN", "Case": "nom|acc", "Gender": "fem", "Number": "plur"},
|
||||
"NOUN__Case=NomAcc|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "nom|acc", "Gender": "fem", "Number": "sing"},
|
||||
"NOUN__Case=NomAcc|Gender=Masc|Number=Plur": {"pos": "NOUN", "Case": "nom|acc", "Gender": "masc", "Number": "plur"},
|
||||
"NOUN__Case=NomAcc|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "nom|acc", "Gender": "masc", "Number": "sing"},
|
||||
"NOUN__Case=Voc|Definite=Def|Gender=Masc|Number=Plur": {"pos": "NOUN", "Case": "voc", "Definite": "def", "Gender": "masc", "Number": "plur"},
|
||||
"NOUN__Case=Voc|Form=Len|Gender=Fem|Number=Sing": {"pos": "NOUN", "Case": "voc", "Gender": "fem", "Number": "sing", "Form": "len"},
|
||||
"NOUN__Case=Voc|Form=Len|Gender=Masc|Number=Plur": {"pos": "NOUN", "Case": "voc", "Gender": "masc", "Number": "plur", "Form": "len"},
|
||||
"NOUN__Case=Voc|Form=Len|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "voc", "Gender": "masc", "Number": "sing", "Form": "len"},
|
||||
"NOUN__Case=Voc|Gender=Masc|Number=Sing": {"pos": "NOUN", "Case": "voc", "Gender": "masc", "Number": "sing"},
|
||||
"NOUN__Degree=Pos": {"pos": "NOUN", "Degree": "pos"},
|
||||
"NOUN__Foreign=Yes": {"pos": "NOUN", "Foreign": "yes"},
|
||||
"NOUN__Form=Ecl|Number=Sing": {"pos": "NOUN", "Number": "sing", "Form": "ecl"},
|
||||
"NOUN__Form=Ecl|VerbForm=Inf": {"pos": "NOUN", "VerbForm": "inf", "Form": "ecl"},
|
||||
"NOUN__Form=Ecl|VerbForm=Vnoun": {"pos": "NOUN", "VerbForm": "vnoun", "Form": "ecl"},
|
||||
"NOUN__Form=HPref|VerbForm=Inf": {"pos": "NOUN", "VerbForm": "inf", "Form": "hpref"},
|
||||
"NOUN__Form=Len|Number=Sing": {"pos": "NOUN", "Number": "sing", "Form": "len"},
|
||||
"NOUN__Form=Len|VerbForm=Inf": {"pos": "NOUN", "VerbForm": "inf", "Form": "len"},
|
||||
"NOUN__Gender=Fem|Number=Sing": {"pos": "NOUN", "Gender": "fem", "Number": "sing"},
|
||||
"NOUN__Number=Sing|PartType=Comp": {"pos": "NOUN", "Number": "sing", "PartType": "comp"},
|
||||
"NOUN__Number=Sing": {"pos": "NOUN", "Number": "sing"},
|
||||
"NOUN___": {"pos": "NOUN"},
|
||||
"NOUN__Reflex=Yes": {"pos": "NOUN", "Reflex": "yes"},
|
||||
"NOUN__VerbForm=Inf": {"pos": "NOUN", "VerbForm": "inf"},
|
||||
"NOUN__VerbForm=Vnoun": {"pos": "NOUN", "VerbForm": "vnoun"},
|
||||
"NUM__Definite=Def|NumType=Card": {"pos": "NUM", "Definite": "def", "NumType": "card"},
|
||||
"NUM__Form=Ecl|NumType=Card": {"pos": "NUM", "NumType": "card", "Form": "ecl"},
|
||||
"NUM__Form=Ecl|NumType=Ord": {"pos": "NUM", "NumType": "ord", "Form": "ecl"},
|
||||
"NUM__Form=HPref|NumType=Card": {"pos": "NUM", "NumType": "card", "Form": "hpref"},
|
||||
"NUM__Form=Len|NumType=Card": {"pos": "NUM", "NumType": "card", "Form": "len"},
|
||||
"NUM__Form=Len|NumType=Ord": {"pos": "NUM", "NumType": "ord", "Form": "len"},
|
||||
"NUM__NumType=Card": {"pos": "NUM", "NumType": "card"},
|
||||
"NUM__NumType=Ord": {"pos": "NUM", "NumType": "ord"},
|
||||
"NUM___": {"pos": "NUM"},
|
||||
"PART__Form=Ecl|PartType=Vb|PronType=Rel": {"pos": "PART", "PronType": "rel", "Form": "ecl", "PartType": "vb"},
|
||||
"PART__Mood=Imp|PartType=Vb|Polarity=Neg": {"pos": "PART", "Mood": "imp", "Polarity": "neg", "PartType": "vb"},
|
||||
"PART__Mood=Imp|PartType=Vb": {"pos": "PART", "Mood": "imp", "PartType": "vb"},
|
||||
"PART__Mood=Int|PartType=Vb|Polarity=Neg": {"pos": "PART", "Polarity": "neg", "Mood": "int", "PartType": "vb"},
|
||||
"PART__PartType=Ad": {"pos": "PART", "PartType": "ad"},
|
||||
"PART__PartType=Cmpl|Polarity=Neg": {"pos": "PART", "Polarity": "neg", "PartType": "cmpl"},
|
||||
"PART__PartType=Cmpl|Polarity=Neg|Tense=Past": {"pos": "PART", "Polarity": "neg", "Tense": "past", "PartType": "cmpl"},
|
||||
"PART__PartType=Cmpl": {"pos": "PART", "PartType": "cmpl"},
|
||||
"PART__PartType=Comp": {"pos": "PART", "PartType": "comp"},
|
||||
"PART__PartType=Cop|PronType=Rel": {"pos": "PART", "PronType": "rel", "PartType": "cop"},
|
||||
"PART__PartType=Deg": {"pos": "PART", "PartType": "deg"},
|
||||
"PART__PartType=Inf": {"pos": "PART", "PartType": "inf"},
|
||||
"PART__PartType=Num": {"pos": "PART", "PartType": "num"},
|
||||
"PART__PartType=Pat": {"pos": "PART", "PartType": "pat"},
|
||||
"PART__PartType=Vb|Polarity=Neg": {"pos": "PART", "Polarity": "neg", "PartType": "vb"},
|
||||
"PART__PartType=Vb|Polarity=Neg|PronType=Rel": {"pos": "PART", "Polarity": "neg", "PronType": "rel", "PartType": "vb"},
|
||||
"PART__PartType=Vb|Polarity=Neg|PronType=Rel|Tense=Past": {"pos": "PART", "Polarity": "neg", "PronType": "rel", "Tense": "past", "PartType": "vb"},
|
||||
"PART__PartType=Vb|Polarity=Neg|Tense=Past": {"pos": "PART", "Polarity": "neg", "Tense": "past", "PartType": "vb"},
|
||||
"PART__PartType=Vb": {"pos": "PART", "PartType": "vb"},
|
||||
"PART__PartType=Vb|PronType=Rel": {"pos": "PART", "PronType": "rel", "PartType": "vb"},
|
||||
"PART__PartType=Vb|PronType=Rel|Tense=Past": {"pos": "PART", "PronType": "rel", "Tense": "past", "PartType": "vb"},
|
||||
"PART__PartType=Vb|Tense=Past": {"pos": "PART", "Tense": "past", "PartType": "vb"},
|
||||
"PART__PartType=Voc": {"pos": "PART", "PartType": "voc"},
|
||||
"PART___": {"pos": "PART"},
|
||||
"PART__PronType=Rel": {"pos": "PART", "PronType": "rel"},
|
||||
"PRON__Form=Len|Number=Sing|Person=2": {"pos": "PRON", "Number": "sing", "Person": 2, "Form": "len"},
|
||||
"PRON__Form=Len|PronType=Ind": {"pos": "PRON", "PronType": "ind", "Form": "len"},
|
||||
"PRON__Gender=Fem|Number=Sing|Person=3": {"pos": "PRON", "Gender": "fem", "Number": "sing", "Person": 3},
|
||||
"PRON__Gender=Masc|Number=Sing|Person=3": {"pos": "PRON", "Gender": "masc", "Number": "sing", "Person": 3},
|
||||
"PRON__Gender=Masc|Number=Sing|Person=3|PronType=Emp": {"pos": "PRON", "Gender": "masc", "Number": "sing", "Person": 3, "PronType": "emp"},
|
||||
"PRON__Gender=Masc|Person=3": {"pos": "PRON", "Gender": "masc", "Person": 3},
|
||||
"PRON__Number=Plur|Person=1": {"pos": "PRON", "Number": "plur", "Person": 1},
|
||||
"PRON__Number=Plur|Person=1|PronType=Emp": {"pos": "PRON", "Number": "plur", "Person": 1, "PronType": "emp"},
|
||||
"PRON__Number=Plur|Person=2": {"pos": "PRON", "Number": "plur", "Person": 2},
|
||||
"PRON__Number=Plur|Person=3": {"pos": "PRON", "Number": "plur", "Person": 3},
|
||||
"PRON__Number=Plur|Person=3|PronType=Emp": {"pos": "PRON", "Number": "plur", "Person": 3, "PronType": "emp"},
|
||||
"PRON__Number=Sing|Person=1": {"pos": "PRON", "Number": "sing", "Person": 1},
|
||||
"PRON__Number=Sing|Person=1|PronType=Emp": {"pos": "PRON", "Number": "sing", "Person": 1, "PronType": "emp"},
|
||||
"PRON__Number=Sing|Person=2": {"pos": "PRON", "Number": "sing", "Person": 2},
|
||||
"PRON__Number=Sing|Person=2|PronType=Emp": {"pos": "PRON", "Number": "sing", "Person": 2, "PronType": "emp"},
|
||||
"PRON__Number=Sing|Person=3": {"pos": "PRON", "Number": "sing", "Person": 3},
|
||||
"PRON__Number=Sing|PronType=Int": {"pos": "PRON", "Number": "sing", "PronType": "int"},
|
||||
"PRON__PronType=Dem": {"pos": "PRON", "PronType": "dem"},
|
||||
"PRON__PronType=Ind": {"pos": "PRON", "PronType": "ind"},
|
||||
"PRON__PronType=Int": {"pos": "PRON", "PronType": "int"},
|
||||
"PRON__Reflex=Yes": {"pos": "PRON", "Reflex": "yes"},
|
||||
"PROPN__Abbr=Yes": {"pos": "PROPN", "Abbr": "yes"},
|
||||
"PROPN__Case=Dat|Gender=Fem|Number=Sing": {"pos": "PROPN", "Case": "dat", "Gender": "fem", "Number": "sing"},
|
||||
"PROPN__Case=Gen|Definite=Def|Gender=Fem|Number=Sing": {"pos": "PROPN", "Case": "gen", "Definite": "def", "Gender": "fem", "Number": "sing"},
|
||||
"PROPN__Case=Gen|Form=Ecl|Gender=Fem|Number=Plur": {"pos": "PROPN", "Case": "gen", "Gender": "fem", "Number": "plur", "Form": "ecl"},
|
||||
"PROPN__Case=Gen|Form=Ecl|Gender=Masc|Number=Plur": {"pos": "PROPN", "Case": "gen", "Gender": "masc", "Number": "plur", "Form": "ecl"},
|
||||
"PROPN__Case=Gen|Form=HPref|Gender=Fem|Number=Sing": {"pos": "PROPN", "Case": "gen", "Gender": "fem", "Number": "sing", "Form": "hpref"},
|
||||
"PROPN__Case=Gen|Form=Len|Gender=Fem|Number=Sing": {"pos": "PROPN", "Case": "gen", "Gender": "fem", "Number": "sing", "Form": "len"},
|
||||
"PROPN__Case=Gen|Form=Len|Gender=Fem": {"pos": "PROPN", "Case": "gen", "Gender": "fem", "Form": "len"},
|
||||
"PROPN__Case=Gen|Form=Len|Gender=Masc|Number=Sing": {"pos": "PROPN", "Case": "gen", "Gender": "masc", "Number": "sing", "Form": "len"},
|
||||
"PROPN__Case=Gen|Form=Len|Gender=Masc": {"pos": "PROPN", "Case": "gen", "Gender": "masc", "Form": "len"},
|
||||
"PROPN__Case=Gen|Gender=Fem|Number=Sing": {"pos": "PROPN", "Case": "gen", "Gender": "fem", "Number": "sing"},
|
||||
"PROPN__Case=Gen|Gender=Fem": {"pos": "PROPN", "Case": "gen", "Gender": "fem"},
|
||||
"PROPN__Case=Gen|Gender=Masc|NounType=Weak|Number=Plur": {"pos": "PROPN", "Case": "gen", "Gender": "masc", "Number": "plur", "NounType": "weak"},
|
||||
"PROPN__Case=Gen|Gender=Masc|Number=Sing": {"pos": "PROPN", "Case": "gen", "Gender": "masc", "Number": "sing"},
|
||||
"PROPN__Case=Gen|Gender=Masc": {"pos": "PROPN", "Case": "gen", "Gender": "masc"},
|
||||
"PROPN__Case=NomAcc|Definite=Def|Gender=Fem|Number=Sing": {"pos": "PROPN", "Case": "nom|acc", "Definite": "def", "Gender": "fem", "Number": "sing"},
|
||||
"PROPN__Case=NomAcc|Definite=Def|Gender=Masc|Number=Plur": {"pos": "PROPN", "Case": "nom|acc", "Definite": "def", "Gender": "masc", "Number": "plur"},
|
||||
"PROPN__Case=NomAcc|Definite=Def|Gender=Masc|Number=Sing": {"pos": "PROPN", "Case": "nom|acc", "Definite": "def", "Gender": "masc", "Number": "sing"},
|
||||
"PROPN__Case=NomAcc|Form=Ecl|Gender=Fem|Number=Sing": {"pos": "PROPN", "Case": "nom|acc", "Gender": "fem", "Number": "sing", "Form": "ecl"},
|
||||
"PROPN__Case=NomAcc|Form=Ecl|Gender=Masc|Number=Sing": {"pos": "PROPN", "Case": "nom|acc", "Gender": "masc", "Number": "sing", "Form": "ecl"},
|
||||
"PROPN__Case=NomAcc|Form=HPref|Gender=Masc|Number=Sing": {"pos": "PROPN", "Case": "nom|acc", "Gender": "masc", "Number": "sing", "Form": "hpref"},
|
||||
"PROPN__Case=NomAcc|Form=Len|Gender=Fem|Number=Sing": {"pos": "PROPN", "Case": "nom|acc", "Gender": "fem", "Number": "sing", "Form": "len"},
|
||||
"PROPN__Case=NomAcc|Form=Len|Gender=Masc|Number=Sing": {"pos": "PROPN", "Case": "nom|acc", "Gender": "masc", "Number": "sing", "Form": "len"},
|
||||
"PROPN__Case=NomAcc|Gender=Fem|Number=Sing": {"pos": "PROPN", "Case": "nom|acc", "Gender": "fem", "Number": "sing"},
|
||||
"PROPN__Case=NomAcc|Gender=Masc|Number=Plur": {"pos": "PROPN", "Case": "nom|acc", "Gender": "masc", "Number": "plur"},
|
||||
"PROPN__Case=NomAcc|Gender=Masc|Number=Sing": {"pos": "PROPN", "Case": "nom|acc", "Gender": "masc", "Number": "sing"},
|
||||
"PROPN__Case=NomAcc|Gender=Masc": {"pos": "PROPN", "Case": "nom|acc", "Gender": "masc"},
|
||||
"PROPN__Case=Voc|Form=Len|Gender=Fem": {"pos": "PROPN", "Case": "voc", "Gender": "fem", "Form": "len"},
|
||||
"PROPN__Case=Voc|Gender=Masc|Number=Sing": {"pos": "PROPN", "Case": "voc", "Gender": "masc", "Number": "sing"},
|
||||
"PROPN__Gender=Masc|Number=Sing": {"pos": "PROPN", "Gender": "masc", "Number": "sing"},
|
||||
"PROPN___": {"pos": "PROPN"},
|
||||
"PUNCT___": {"pos": "PUNCT"},
|
||||
"SCONJ___": {"pos": "SCONJ"},
|
||||
"SCONJ__Tense=Past|VerbForm=Cop": {"pos": "SCONJ", "Tense": "past", "VerbForm": "cop"},
|
||||
"SCONJ__VerbForm=Cop": {"pos": "SCONJ", "VerbForm": "cop"},
|
||||
"SYM__Abbr=Yes": {"pos": "SYM", "Abbr": "yes"},
|
||||
"VERB__Case=NomAcc|Gender=Masc|Mood=Ind|Number=Sing|Tense=Pres": {"pos": "VERB", "Case": "nom|acc", "Gender": "masc", "Mood": "ind", "Number": "sing", "Tense": "pres"},
|
||||
"VERB__Dialect=Munster|Form=Len|Mood=Ind|Tense=Past": {"pos": "VERB", "Mood": "ind", "Tense": "past", "Dialect": "munster", "Form": "len"},
|
||||
"VERB__Foreign=Yes": {"pos": "VERB", "Foreign": "yes"},
|
||||
"VERB__Form=Ecl|Mood=Cnd|Number=Sing|Person=1": {"pos": "VERB", "Mood": "cnd", "Number": "sing", "Person": 1, "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Cnd|Polarity=Neg": {"pos": "VERB", "Mood": "cnd", "Polarity": "neg", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Cnd": {"pos": "VERB", "Mood": "cnd", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Cnd|Voice=Auto": {"pos": "VERB", "Mood": "cnd", "Form": "ecl", "Voice": "auto"},
|
||||
"VERB__Form=Ecl|Mood=Imp|Number=Sing|Person=1|Tense=Past": {"pos": "VERB", "Mood": "imp", "Number": "sing", "Person": 1, "Tense": "past", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Imp|Tense=Past": {"pos": "VERB", "Mood": "imp", "Tense": "past", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Ind|Number=Plur|Person=1|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Number": "plur", "Person": 1, "Tense": "pres", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Ind|Number=Sing|Person=1|Tense=Past": {"pos": "VERB", "Mood": "ind", "Number": "sing", "Person": 1, "Tense": "past", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Ind|Number=Sing|Person=1|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Number": "sing", "Person": 1, "Tense": "pres", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Ind|Polarity=Neg|Tense=Fut": {"pos": "VERB", "Mood": "ind", "Polarity": "neg", "Tense": "fut", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Ind|Polarity=Neg|Tense=Fut|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Polarity": "neg", "Tense": "fut", "Form": "ecl", "Voice": "auto"},
|
||||
"VERB__Form=Ecl|Mood=Ind|Polarity=Neg|Tense=Past": {"pos": "VERB", "Mood": "ind", "Polarity": "neg", "Tense": "past", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Ind|Polarity=Neg|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Polarity": "neg", "Tense": "pres", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Ind|Tense=Fut": {"pos": "VERB", "Mood": "ind", "Tense": "fut", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Ind|Tense=Fut|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Tense": "fut", "Form": "ecl", "Voice": "auto"},
|
||||
"VERB__Form=Ecl|Mood=Ind|Tense=Past": {"pos": "VERB", "Mood": "ind", "Tense": "past", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Ind|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Tense": "pres", "Form": "ecl"},
|
||||
"VERB__Form=Ecl|Mood=Ind|Tense=Pres|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Tense": "pres", "Form": "ecl", "Voice": "auto"},
|
||||
"VERB__Form=Ecl|Mood=Sub|Tense=Pres": {"pos": "VERB", "Mood": "sub", "Tense": "pres", "Form": "ecl"},
|
||||
"VERB__Form=Ecl": {"pos": "VERB", "Form": "ecl"},
|
||||
"VERB__Form=Emp|Mood=Ind|Number=Plur|Person=1|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Number": "plur", "Person": 1, "Tense": "pres", "Form": "emp"},
|
||||
"VERB__Form=Emp|Mood=Ind|Number=Sing|Person=1|PronType=Rel|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Number": "sing", "Person": 1, "PronType": "rel", "Tense": "pres", "Form": "emp"},
|
||||
"VERB__Form=Emp|Mood=Ind|Number=Sing|Person=1|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Number": "sing", "Person": 1, "Tense": "pres", "Form": "emp"},
|
||||
"VERB__Form=Len|Mood=Cnd|Number=Plur|Person=3": {"pos": "VERB", "Mood": "cnd", "Number": "plur", "Person": 3, "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Cnd|Number=Sing|Person=1": {"pos": "VERB", "Mood": "cnd", "Number": "sing", "Person": 1, "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Cnd|Number=Sing|Person=2": {"pos": "VERB", "Mood": "cnd", "Number": "sing", "Person": 2, "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Cnd|Polarity=Neg": {"pos": "VERB", "Mood": "cnd", "Polarity": "neg", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Cnd": {"pos": "VERB", "Mood": "cnd", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Cnd|Voice=Auto": {"pos": "VERB", "Mood": "cnd", "Form": "len", "Voice": "auto"},
|
||||
"VERB__Form=Len|Mood=Imp|Number=Plur|Person=3|Tense=Past": {"pos": "VERB", "Mood": "imp", "Number": "plur", "Person": 3, "Tense": "past", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Imp|Tense=Past": {"pos": "VERB", "Mood": "imp", "Tense": "past", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Imp|Tense=Past|Voice=Auto": {"pos": "VERB", "Mood": "imp", "Tense": "past", "Form": "len", "Voice": "auto"},
|
||||
"VERB__Form=Len|Mood=Imp|Voice=Auto": {"pos": "VERB", "Mood": "imp", "Form": "len", "Voice": "auto"},
|
||||
"VERB__Form=Len|Mood=Ind|Number=Plur|Person=1|Tense=Fut": {"pos": "VERB", "Mood": "ind", "Number": "plur", "Person": 1, "Tense": "fut", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Ind|Number=Plur|Person=1|Tense=Past": {"pos": "VERB", "Mood": "ind", "Number": "plur", "Person": 1, "Tense": "past", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Ind|Number=Plur|Person=3|Tense=Past": {"pos": "VERB", "Mood": "ind", "Number": "plur", "Person": 3, "Tense": "past", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Ind|Number=Sing|Person=1|Polarity=Neg|Tense=Past": {"pos": "VERB", "Mood": "ind", "Number": "sing", "Person": 1, "Polarity": "neg", "Tense": "past", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Ind|Number=Sing|Person=1|Polarity=Neg|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Number": "sing", "Person": 1, "Polarity": "neg", "Tense": "pres", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Ind|Number=Sing|Person=1|Tense=Past": {"pos": "VERB", "Mood": "ind", "Number": "sing", "Person": 1, "Tense": "past", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Ind|Polarity=Neg|Tense=Fut": {"pos": "VERB", "Mood": "ind", "Polarity": "neg", "Tense": "fut", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Ind|Polarity=Neg|Tense=Fut|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Polarity": "neg", "Tense": "fut", "Form": "len", "Voice": "auto"},
|
||||
"VERB__Form=Len|Mood=Ind|Polarity=Neg|Tense=Past": {"pos": "VERB", "Mood": "ind", "Polarity": "neg", "Tense": "past", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Ind|Polarity=Neg|Tense=Past|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Polarity": "neg", "Tense": "past", "Form": "len", "Voice": "auto"},
|
||||
"VERB__Form=Len|Mood=Ind|Polarity=Neg|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Polarity": "neg", "Tense": "pres", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Ind|Polarity=Neg|Tense=Pres|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Polarity": "neg", "Tense": "pres", "Form": "len", "Voice": "auto"},
|
||||
"VERB__Form=Len|Mood=Ind|Tense=Fut": {"pos": "VERB", "Mood": "ind", "Tense": "fut", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Ind|Tense=Fut|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Tense": "fut", "Form": "len", "Voice": "auto"},
|
||||
"VERB__Form=Len|Mood=Ind|Tense=Past": {"pos": "VERB", "Mood": "ind", "Tense": "past", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Ind|Tense=Past|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Tense": "past", "Form": "len", "Voice": "auto"},
|
||||
"VERB__Form=Len|Mood=Ind|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Tense": "pres", "Form": "len"},
|
||||
"VERB__Form=Len|Mood=Ind|Tense=Pres|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Tense": "pres", "Form": "len", "Voice": "auto"},
|
||||
"VERB__Form=Len|Mood=Sub|Polarity=Neg|Tense=Pres": {"pos": "VERB", "Mood": "sub", "Polarity": "neg", "Tense": "pres", "Form": "len"},
|
||||
"VERB__Form=Len|Polarity=Neg": {"pos": "VERB", "Polarity": "neg", "Form": "len"},
|
||||
"VERB__Form=Len": {"pos": "VERB", "Form": "len"},
|
||||
"VERB__Mood=Cnd|Number=Plur|Person=3": {"pos": "VERB", "Mood": "cnd", "Number": "plur", "Person": 3},
|
||||
"VERB__Mood=Cnd|Number=Sing|Person=1": {"pos": "VERB", "Mood": "cnd", "Number": "sing", "Person": 1},
|
||||
"VERB__Mood=Cnd": {"pos": "VERB", "Mood": "cnd"},
|
||||
"VERB__Mood=Cnd|Voice=Auto": {"pos": "VERB", "Mood": "cnd", "Voice": "auto"},
|
||||
"VERB__Mood=Imp|Number=Plur|Person=1|Polarity=Neg": {"pos": "VERB", "Mood": "imp", "Number": "plur", "Person": 1, "Polarity": "neg"},
|
||||
"VERB__Mood=Imp|Number=Plur|Person=1": {"pos": "VERB", "Mood": "imp", "Number": "plur", "Person": 1},
|
||||
"VERB__Mood=Imp|Number=Plur|Person=2": {"pos": "VERB", "Mood": "imp", "Number": "plur", "Person": 2},
|
||||
"VERB__Mood=Imp|Number=Sing|Person=2": {"pos": "VERB", "Mood": "imp", "Number": "sing", "Person": 2},
|
||||
"VERB__Mood=Imp|Tense=Past": {"pos": "VERB", "Mood": "imp", "Tense": "past"},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=1|Tense=Past": {"pos": "VERB", "Mood": "ind", "Number": "plur", "Person": 1, "Tense": "past"},
|
||||
"VERB__Mood=Ind|Number=Plur|Person=1|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Number": "plur", "Person": 1, "Tense": "pres"},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=1|Tense=Past": {"pos": "VERB", "Mood": "ind", "Number": "sing", "Person": 1, "Tense": "past"},
|
||||
"VERB__Mood=Ind|Number=Sing|Person=1|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Number": "sing", "Person": 1, "Tense": "pres"},
|
||||
"VERB__Mood=Ind|Polarity=Neg|Tense=Past|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Polarity": "neg", "Tense": "past", "Voice": "auto"},
|
||||
"VERB__Mood=Ind|Polarity=Neg|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Polarity": "neg", "Tense": "pres"},
|
||||
"VERB__Mood=Ind|PronType=Rel|Tense=Fut": {"pos": "VERB", "Mood": "ind", "PronType": "rel", "Tense": "fut"},
|
||||
"VERB__Mood=Ind|PronType=Rel|Tense=Pres": {"pos": "VERB", "Mood": "ind", "PronType": "rel", "Tense": "pres"},
|
||||
"VERB__Mood=Ind|Tense=Fut": {"pos": "VERB", "Mood": "ind", "Tense": "fut"},
|
||||
"VERB__Mood=Ind|Tense=Fut|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Tense": "fut", "Voice": "auto"},
|
||||
"VERB__Mood=Ind|Tense=Past": {"pos": "VERB", "Mood": "ind", "Tense": "past"},
|
||||
"VERB__Mood=Ind|Tense=Past|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Tense": "past", "Voice": "auto"},
|
||||
"VERB__Mood=Ind|Tense=Pres": {"pos": "VERB", "Mood": "ind", "Tense": "pres"},
|
||||
"VERB__Mood=Ind|Tense=Pres|Voice=Auto": {"pos": "VERB", "Mood": "ind", "Tense": "pres", "Voice": "auto"},
|
||||
"VERB___": {"pos": "VERB"},
|
||||
"X__Abbr=Yes": {"pos": "X", "Abbr": "yes"},
|
||||
"X__Case=NomAcc|Foreign=Yes|Gender=Fem|Number=Sing": {"pos": "X", "Case": "nom|acc", "Gender": "fem", "Number": "sing", "Foreign": "yes"},
|
||||
"X__Definite=Def|Dialect=Ulster": {"pos": "X", "Definite": "def", "Dialect": "ulster"},
|
||||
"X__Dialect=Munster|Form=Len|Mood=Ind|Number=Sing|Person=1|Tense=Past": {"pos": "X", "Mood": "ind", "Number": "sing", "Person": 1, "Tense": "past", "Dialect": "munster", "Form": "len"},
|
||||
"X__Dialect=Munster|Mood=Imp|Number=Sing|Person=2|Polarity=Neg": {"pos": "X", "Mood": "imp", "Number": "sing", "Person": 2, "Polarity": "neg", "Dialect": "munster"},
|
||||
"X__Dialect=Munster|Mood=Ind|Tense=Past|Voice=Auto": {"pos": "X", "Mood": "ind", "Tense": "past", "Dialect": "munster", "Voice": "auto"},
|
||||
"X__Dialect=Munster": {"pos": "X", "Dialect": "munster"},
|
||||
"X__Dialect=Munster|PronType=Dem": {"pos": "X", "PronType": "dem", "Dialect": "munster"},
|
||||
"X__Dialect=Ulster|Gender=Masc|Number=Sing|Person=3": {"pos": "X", "Gender": "masc", "Number": "sing", "Person": 3, "Dialect": "ulster"},
|
||||
"X__Dialect=Ulster|PartType=Vb|Polarity=Neg": {"pos": "X", "Polarity": "neg", "Dialect": "ulster", "PartType": "vb"},
|
||||
"X__Dialect=Ulster|VerbForm=Cop": {"pos": "X", "Dialect": "ulster", "VerbForm": "cop"},
|
||||
"X__Foreign=Yes": {"pos": "X", "Foreign": "yes"},
|
||||
"X___": {"pos": "X"}
|
||||
}
|
||||
# fmt: on
|
||||
|
|
@ -1,6 +1,5 @@
|
|||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .tag_map import TAG_MAP
|
||||
|
||||
from ...attrs import LANG
|
||||
from ...language import Language
|
||||
|
|
@ -12,7 +11,6 @@ class ArmenianDefaults(Language.Defaults):
|
|||
|
||||
lex_attr_getters.update(LEX_ATTRS)
|
||||
stop_words = STOP_WORDS
|
||||
tag_map = TAG_MAP
|
||||
|
||||
|
||||
class Armenian(Language):
|
||||
|
|
|
|||
File diff suppressed because it is too large
Load Diff
|
|
@ -3,7 +3,6 @@ from .punctuation import TOKENIZER_SUFFIXES, TOKENIZER_PREFIXES, TOKENIZER_INFIX
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .syntax_iterators import SYNTAX_ITERATORS
|
||||
from .tag_map import TAG_MAP
|
||||
|
||||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
from ...language import Language
|
||||
|
|
@ -21,7 +20,6 @@ class IndonesianDefaults(Language.Defaults):
|
|||
suffixes = TOKENIZER_SUFFIXES
|
||||
infixes = TOKENIZER_INFIXES
|
||||
syntax_iterators = SYNTAX_ITERATORS
|
||||
tag_map = TAG_MAP
|
||||
|
||||
|
||||
class Indonesian(Language):
|
||||
|
|
|
|||
|
|
@ -1,92 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, ADJ, CCONJ, NUM, DET, ADV, ADP, X, VERB
|
||||
from ...symbols import NOUN, PRON, AUX, SCONJ, INTJ, PART, PROPN
|
||||
|
||||
|
||||
# POS explanations for indonesian available from https://www.aclweb.org/anthology/Y12-1014
|
||||
TAG_MAP = {
|
||||
"NSD": {POS: NOUN},
|
||||
"Z--": {POS: PUNCT},
|
||||
"VSA": {POS: VERB},
|
||||
"CC-": {POS: NUM},
|
||||
"R--": {POS: ADP},
|
||||
"D--": {POS: ADV},
|
||||
"ASP": {POS: ADJ},
|
||||
"S--": {POS: SCONJ},
|
||||
"VSP": {POS: VERB},
|
||||
"H--": {POS: CCONJ},
|
||||
"F--": {POS: X},
|
||||
"B--": {POS: DET},
|
||||
"CO-": {POS: NUM},
|
||||
"G--": {POS: ADV},
|
||||
"PS3": {POS: PRON},
|
||||
"W--": {POS: ADV},
|
||||
"O--": {POS: AUX},
|
||||
"PP1": {POS: PRON},
|
||||
"ASS": {POS: ADJ},
|
||||
"PS1": {POS: PRON},
|
||||
"APP": {POS: ADJ},
|
||||
"CD-": {POS: NUM},
|
||||
"VPA": {POS: VERB},
|
||||
"VPP": {POS: VERB},
|
||||
"X--": {POS: X},
|
||||
"CO-+PS3": {POS: NUM},
|
||||
"NSD+PS3": {POS: NOUN},
|
||||
"ASP+PS3": {POS: ADJ},
|
||||
"M--": {POS: AUX},
|
||||
"VSA+PS3": {POS: VERB},
|
||||
"R--+PS3": {POS: ADP},
|
||||
"W--+T--": {POS: ADV},
|
||||
"PS2": {POS: PRON},
|
||||
"NSD+PS1": {POS: NOUN},
|
||||
"PP3": {POS: PRON},
|
||||
"VSA+T--": {POS: VERB},
|
||||
"D--+T--": {POS: ADV},
|
||||
"VSP+PS3": {POS: VERB},
|
||||
"F--+PS3": {POS: X},
|
||||
"M--+T--": {POS: AUX},
|
||||
"F--+T--": {POS: X},
|
||||
"PUNCT": {POS: PUNCT},
|
||||
"PROPN": {POS: PROPN},
|
||||
"I--": {POS: INTJ},
|
||||
"S--+PS3": {POS: SCONJ},
|
||||
"ASP+T--": {POS: ADJ},
|
||||
"CC-+PS3": {POS: NUM},
|
||||
"NSD+PS2": {POS: NOUN},
|
||||
"B--+T--": {POS: DET},
|
||||
"H--+T--": {POS: CCONJ},
|
||||
"VSA+PS2": {POS: VERB},
|
||||
"NSF": {POS: NOUN},
|
||||
"PS1+VSA": {POS: PRON},
|
||||
"NPD": {POS: NOUN},
|
||||
"PP2": {POS: PRON},
|
||||
"VSA+PS1": {POS: VERB},
|
||||
"T--": {POS: PART},
|
||||
"NSM": {POS: NOUN},
|
||||
"NUM": {POS: NUM},
|
||||
"ASP+PS2": {POS: ADJ},
|
||||
"G--+T--": {POS: PART},
|
||||
"D--+PS3": {POS: ADV},
|
||||
"R--+PS2": {POS: ADP},
|
||||
"NSM+PS3": {POS: NOUN},
|
||||
"VSP+T--": {POS: VERB},
|
||||
"M--+PS3": {POS: AUX},
|
||||
"ASS+PS3": {POS: ADJ},
|
||||
"G--+PS3": {POS: PART},
|
||||
"F--+PS1": {POS: X},
|
||||
"NSD+T--": {POS: NOUN},
|
||||
"PP1+T--": {POS: PRON},
|
||||
"B--+PS3": {POS: DET},
|
||||
"NOUN": {POS: NOUN},
|
||||
"NPD+PS3": {POS: NOUN},
|
||||
"R--+PS1": {POS: ADP},
|
||||
"F--+PS2": {POS: X},
|
||||
"CD-+PS3": {POS: NUM},
|
||||
"PS1+VSA+T--": {POS: VERB},
|
||||
"PS2+VSA": {POS: VERB},
|
||||
"VERB": {POS: VERB},
|
||||
"CC-+T--": {POS: NUM},
|
||||
"NPD+PS2": {POS: NOUN},
|
||||
"D--+PS2": {POS: ADV},
|
||||
"PP3+T--": {POS: PRON},
|
||||
"X": {POS: X},
|
||||
}
|
||||
|
|
@ -1,5 +1,4 @@
|
|||
from .stop_words import STOP_WORDS
|
||||
from .tag_map import TAG_MAP
|
||||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .punctuation import TOKENIZER_PREFIXES, TOKENIZER_INFIXES
|
||||
|
||||
|
|
@ -18,7 +17,6 @@ class ItalianDefaults(Language.Defaults):
|
|||
)
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
stop_words = STOP_WORDS
|
||||
tag_map = TAG_MAP
|
||||
prefixes = TOKENIZER_PREFIXES
|
||||
infixes = TOKENIZER_INFIXES
|
||||
|
||||
|
|
|
|||
|
|
@ -1,320 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, SYM, ADJ, NUM, DET, ADV, ADP, X, VERB
|
||||
from ...symbols import NOUN, PROPN, PART, INTJ, SPACE, PRON, SCONJ, AUX, CONJ
|
||||
|
||||
|
||||
TAG_MAP = {
|
||||
"AP__Gender=Fem|Number=Plur|Poss=Yes|PronType=Prs": {POS: DET},
|
||||
"AP__Gender=Fem|Number=Sing|Poss=Yes|PronType=Prs": {POS: DET},
|
||||
"AP__Gender=Masc|Number=Plur|Poss=Yes|PronType=Prs": {POS: DET},
|
||||
"AP__Gender=Masc|Number=Sing|Poss=Yes|PronType=Prs": {POS: DET},
|
||||
"AP__Gender=Masc|Poss=Yes|PronType=Prs": {POS: DET},
|
||||
"AP__Number=Sing|Poss=Yes|PronType=Prs": {POS: DET},
|
||||
"AP__Poss=Yes|PronType=Prs": {POS: DET},
|
||||
"A__Degree=Abs|Gender=Fem|Number=Plur": {POS: ADJ},
|
||||
"A__Degree=Abs|Gender=Fem|Number=Sing": {POS: ADJ},
|
||||
"A__Degree=Abs|Gender=Masc|Number=Plur": {POS: ADJ},
|
||||
"A__Degree=Abs|Gender=Masc|Number=Sing": {POS: ADJ},
|
||||
"A__Degree=Cmp": {POS: ADJ},
|
||||
"A__Degree=Cmp|Number=Plur": {POS: ADJ},
|
||||
"A__Degree=Cmp|Number=Sing": {POS: ADJ},
|
||||
"A__Gender=Fem|Number=Plur": {POS: ADJ},
|
||||
"A__Gender=Fem|Number=Sing": {POS: ADJ},
|
||||
"A__Gender=Fem|Number=Sing|Poss=Yes|PronType=Prs": {POS: ADJ},
|
||||
"A__Gender=Masc": {POS: ADJ},
|
||||
"A__Gender=Masc|Number=Plur": {POS: ADJ},
|
||||
"A__Gender=Masc|Number=Sing": {POS: ADJ},
|
||||
"A__Number=Plur": {POS: ADJ},
|
||||
"A__Number=Sing": {POS: ADJ},
|
||||
"A___": {POS: ADJ},
|
||||
"BN__PronType=Neg": {POS: ADV},
|
||||
"B__Degree=Abs": {POS: ADV},
|
||||
"B__Degree=Abs|Gender=Masc|Number=Sing": {POS: ADV},
|
||||
"B___": {POS: ADV},
|
||||
"CC___": {POS: CONJ},
|
||||
"CS___": {POS: SCONJ},
|
||||
"DD__Gender=Fem|Number=Plur|PronType=Dem": {POS: DET},
|
||||
"DD__Gender=Fem|Number=Sing|PronType=Dem": {POS: DET},
|
||||
"DD__Gender=Masc|Number=Plur|PronType=Dem": {POS: DET},
|
||||
"DD__Gender=Masc|Number=Sing|PronType=Dem": {POS: DET},
|
||||
"DD__Gender=Masc|PronType=Dem": {POS: DET},
|
||||
"DD__Number=Plur|PronType=Dem": {POS: DET},
|
||||
"DD__Number=Sing|PronType=Dem": {POS: DET},
|
||||
"DE__PronType=Exc": {POS: DET},
|
||||
"DI__Definite=Def|Gender=Fem|Number=Plur|PronType=Art": {POS: DET},
|
||||
"DI__Gender=Fem|Number=Plur": {POS: DET},
|
||||
"DI__Gender=Fem|Number=Plur|PronType=Ind": {POS: DET},
|
||||
"DI__Gender=Fem|Number=Sing|PronType=Ind": {POS: DET},
|
||||
"DI__Gender=Masc|Number=Plur": {POS: DET},
|
||||
"DI__Gender=Masc|Number=Plur|PronType=Ind": {POS: DET},
|
||||
"DI__Gender=Masc|Number=Sing|PronType=Ind": {POS: DET},
|
||||
"DI__Number=Sing|PronType=Art": {POS: DET},
|
||||
"DI__Number=Sing|PronType=Ind": {POS: DET},
|
||||
"DI__PronType=Ind": {POS: DET},
|
||||
"DQ__Gender=Fem|Number=Plur|PronType=Int": {POS: DET},
|
||||
"DQ__Gender=Fem|Number=Sing|PronType=Int": {POS: DET},
|
||||
"DQ__Gender=Masc|Number=Plur|PronType=Int": {POS: DET},
|
||||
"DQ__Gender=Masc|Number=Sing|PronType=Int": {POS: DET},
|
||||
"DQ__Number=Plur|PronType=Int": {POS: DET},
|
||||
"DQ__Number=Sing|PronType=Int": {POS: DET},
|
||||
"DQ__PronType=Int": {POS: DET},
|
||||
"DQ___": {POS: DET},
|
||||
"DR__Number=Plur|PronType=Rel": {POS: DET},
|
||||
"DR__PronType=Rel": {POS: DET},
|
||||
"E__Gender=Masc|Number=Sing": {POS: ADP},
|
||||
"E___": {POS: ADP},
|
||||
"FB___": {POS: PUNCT},
|
||||
"FC___": {POS: PUNCT},
|
||||
"FF___": {POS: PUNCT},
|
||||
"FS___": {POS: PUNCT},
|
||||
"I__Polarity=Neg": {POS: INTJ},
|
||||
"I__Polarity=Pos": {POS: INTJ},
|
||||
"I___": {POS: INTJ},
|
||||
"NO__Gender=Fem|Number=Plur|NumType=Ord": {POS: ADJ},
|
||||
"NO__Gender=Fem|Number=Sing|NumType=Ord": {POS: ADJ},
|
||||
"NO__Gender=Masc|Number=Plur": {POS: ADJ},
|
||||
"NO__Gender=Masc|Number=Plur|NumType=Ord": {POS: ADJ},
|
||||
"NO__Gender=Masc|Number=Sing|NumType=Ord": {POS: ADJ},
|
||||
"NO__NumType=Ord": {POS: ADJ},
|
||||
"NO__Number=Sing|NumType=Ord": {POS: ADJ},
|
||||
"NO___": {POS: ADJ},
|
||||
"N__Gender=Masc|Number=Sing": {POS: NUM},
|
||||
"N__NumType=Card": {POS: NUM},
|
||||
"N__NumType=Range": {POS: NUM},
|
||||
"N___": {POS: NUM},
|
||||
"PART___": {POS: PART},
|
||||
"PC__Clitic=Yes|Definite=Def|Gender=Fem|Number=Plur|PronType=Art": {POS: PRON},
|
||||
"PC__Clitic=Yes|Gender=Fem|Number=Plur|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Gender=Fem|Number=Plur|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Gender=Fem|Number=Sing|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Gender=Fem|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Gender=Masc|Number=Plur|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Gender=Masc|Number=Sing|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Gender=Masc|Number=Sing|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Number=Plur|Person=1|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Number=Plur|Person=2|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Number=Plur|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Number=Plur|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Number=Sing|Person=1|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Number=Sing|Person=2|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Number=Sing|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PC__Clitic=Yes|PronType=Prs": {POS: PRON},
|
||||
"PD__Gender=Fem|Number=Plur|PronType=Dem": {POS: PRON},
|
||||
"PD__Gender=Fem|Number=Sing|PronType=Dem": {POS: PRON},
|
||||
"PD__Gender=Masc|Number=Plur|PronType=Dem": {POS: PRON},
|
||||
"PD__Gender=Masc|Number=Sing|PronType=Dem": {POS: PRON},
|
||||
"PD__Number=Plur|PronType=Dem": {POS: PRON},
|
||||
"PD__Number=Sing|PronType=Dem": {POS: PRON},
|
||||
"PD__PronType=Dem": {POS: PRON},
|
||||
"PE__Gender=Fem|Number=Plur|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PE__Gender=Fem|Number=Sing|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PE__Gender=Masc|Number=Plur|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PE__Gender=Masc|Number=Sing|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PE__Number=Plur|Person=1|PronType=Prs": {POS: PRON},
|
||||
"PE__Number=Plur|Person=2|PronType=Prs": {POS: PRON},
|
||||
"PE__Number=Plur|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PE__Number=Sing|Person=1|PronType=Prs": {POS: PRON},
|
||||
"PE__Number=Sing|Person=2|PronType=Prs": {POS: PRON},
|
||||
"PE__Number=Sing|Person=3|PronType=Prs": {POS: PRON},
|
||||
"PE__Person=3|PronType=Prs": {POS: PRON},
|
||||
"PE__PronType=Prs": {POS: PRON},
|
||||
"PI__Gender=Fem|Number=Plur|PronType=Ind": {POS: PRON},
|
||||
"PI__Gender=Fem|Number=Sing|PronType=Ind": {POS: PRON},
|
||||
"PI__Gender=Masc|Number=Plur|PronType=Ind": {POS: PRON},
|
||||
"PI__Gender=Masc|Number=Sing": {POS: PRON},
|
||||
"PI__Gender=Masc|Number=Sing|PronType=Ind": {POS: PRON},
|
||||
"PI__Number=Plur|PronType=Ind": {POS: PRON},
|
||||
"PI__Number=Sing|PronType=Ind": {POS: PRON},
|
||||
"PI__PronType=Ind": {POS: PRON},
|
||||
"PP__Gender=Fem|Number=Sing|Poss=Yes|PronType=Prs": {POS: PRON},
|
||||
"PP__Gender=Masc|Number=Plur|Poss=Yes|PronType=Prs": {POS: PRON},
|
||||
"PP__Gender=Masc|Number=Sing|Poss=Yes|PronType=Prs": {POS: PRON},
|
||||
"PP__Number=Plur|Poss=Yes|PronType=Prs": {POS: PRON},
|
||||
"PP__Number=Sing|Poss=Yes|PronType=Prs": {POS: PRON},
|
||||
"PQ__Gender=Fem|Number=Plur|PronType=Int": {POS: PRON},
|
||||
"PQ__Gender=Fem|Number=Sing|PronType=Int": {POS: PRON},
|
||||
"PQ__Gender=Masc|Number=Plur|PronType=Int": {POS: PRON},
|
||||
"PQ__Gender=Masc|Number=Sing|PronType=Int": {POS: PRON},
|
||||
"PQ__Number=Plur|PronType=Int": {POS: PRON},
|
||||
"PQ__Number=Sing|PronType=Int": {POS: PRON},
|
||||
"PQ__PronType=Int": {POS: PRON},
|
||||
"PR__Gender=Masc|Number=Plur|PronType=Rel": {POS: PRON},
|
||||
"PR__Gender=Masc|Number=Sing|PronType=Rel": {POS: PRON},
|
||||
"PR__Gender=Masc|PronType=Rel": {POS: PRON},
|
||||
"PR__Number=Plur|PronType=Rel": {POS: PRON},
|
||||
"PR__Number=Sing|PronType=Rel": {POS: PRON},
|
||||
"PR__Person=3|PronType=Rel": {POS: PRON},
|
||||
"PR__PronType=Rel": {POS: PRON},
|
||||
"RD__Definite=Def": {POS: DET},
|
||||
"RD__Definite=Def|Gender=Fem": {POS: DET},
|
||||
"RD__Definite=Def|Gender=Fem|Number=Plur|PronType=Art": {POS: DET},
|
||||
"RD__Definite=Def|Gender=Fem|Number=Sing|PronType=Art": {POS: DET},
|
||||
"RD__Definite=Def|Gender=Masc|Number=Plur|PronType=Art": {POS: DET},
|
||||
"RD__Definite=Def|Gender=Masc|Number=Sing|PronType=Art": {POS: DET},
|
||||
"RD__Definite=Def|Number=Plur|PronType=Art": {POS: DET},
|
||||
"RD__Definite=Def|Number=Sing|PronType=Art": {POS: DET},
|
||||
"RD__Definite=Def|PronType=Art": {POS: DET},
|
||||
"RD__Gender=Fem|Number=Sing": {POS: DET},
|
||||
"RD__Gender=Masc|Number=Sing": {POS: DET},
|
||||
"RD__Number=Sing": {POS: DET},
|
||||
"RD__Number=Sing|PronType=Art": {POS: DET},
|
||||
"RI__Definite=Ind|Gender=Fem|Number=Plur|PronType=Art": {POS: DET},
|
||||
"RI__Definite=Ind|Gender=Fem|Number=Sing|PronType=Art": {POS: DET},
|
||||
"RI__Definite=Ind|Gender=Masc|Number=Plur|PronType=Art": {POS: DET},
|
||||
"RI__Definite=Ind|Gender=Masc|Number=Sing|PronType=Art": {POS: DET},
|
||||
"RI__Definite=Ind|Number=Sing|PronType=Art": {POS: DET},
|
||||
"RI__Definite=Ind|PronType=Art": {POS: DET},
|
||||
"SP__Gender=Fem|Number=Plur": {POS: PROPN},
|
||||
"SP__NumType=Card": {POS: PROPN},
|
||||
"SP___": {POS: PROPN},
|
||||
"SW__Foreign=Yes": {POS: X},
|
||||
"SW__Foreign=Yes|Gender=Masc": {POS: X},
|
||||
"SW__Foreign=Yes|Number=Sing": {POS: X},
|
||||
"SYM___": {POS: SYM},
|
||||
"S__Gender=Fem": {POS: NOUN},
|
||||
"S__Gender=Fem|Number=Plur": {POS: NOUN},
|
||||
"S__Gender=Fem|Number=Sing": {POS: NOUN},
|
||||
"S__Gender=Masc": {POS: NOUN},
|
||||
"S__Gender=Masc|Number=Plur": {POS: NOUN},
|
||||
"S__Gender=Masc|Number=Sing": {POS: NOUN},
|
||||
"S__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part": {POS: NOUN},
|
||||
"S__Number=Plur": {POS: NOUN},
|
||||
"S__Number=Sing": {POS: NOUN},
|
||||
"S___": {POS: NOUN},
|
||||
"Sw___": {POS: X},
|
||||
"T__Gender=Fem|Number=Plur|PronType=Tot": {POS: DET},
|
||||
"T__Gender=Fem|Number=Sing": {POS: DET},
|
||||
"T__Gender=Fem|Number=Sing|PronType=Tot": {POS: DET},
|
||||
"T__Gender=Masc|Number=Plur|PronType=Tot": {POS: DET},
|
||||
"T__Gender=Masc|Number=Sing|PronType=Tot": {POS: DET},
|
||||
"T__Number=Plur|PronType=Tot": {POS: DET},
|
||||
"T__PronType=Tot": {POS: DET},
|
||||
"VA__Gender=Fem|Number=Plur|Tense=Past|VerbForm=Part": {POS: AUX},
|
||||
"VA__Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part": {POS: AUX},
|
||||
"VA__Gender=Masc|Number=Plur|Tense=Past|VerbForm=Part": {POS: AUX},
|
||||
"VA__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part": {POS: AUX},
|
||||
"VA__Mood=Cnd|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Cnd|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Cnd|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Cnd|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Plur|Person=2|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Plur|Person=2|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Plur|Person=3|Tense=Past|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Sing|Person=1|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Sing|Person=1|Tense=Past|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Sing|Person=2|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Sing|Person=3|Tense=Past|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Sub|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Sub|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Sub|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Sub|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Sub|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VA__Mood=Sub|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VA__VerbForm=Ger": {POS: AUX},
|
||||
"VA__VerbForm=Inf": {POS: AUX},
|
||||
"VM__Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part": {POS: AUX},
|
||||
"VM__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part": {POS: AUX},
|
||||
"VM__Mood=Cnd|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Cnd|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Cnd|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Cnd|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Cnd|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Cnd|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Imp|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Imp|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Plur|Person=2|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Plur|Person=3|Tense=Past|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Sing|Person=2|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Sing|Person=2|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Sing|Person=3|Tense=Past|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Sub|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Sub|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Sub|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Sub|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Sub|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {POS: AUX},
|
||||
"VM__Mood=Sub|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: AUX},
|
||||
"VM__VerbForm=Ger": {POS: AUX},
|
||||
"VM__VerbForm=Inf": {POS: AUX},
|
||||
"V__Gender=Fem|Number=Plur|Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"V__Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"V__Gender=Masc|Number=Plur|Tense=Past|VerbForm=Fin": {POS: VERB},
|
||||
"V__Gender=Masc|Number=Plur|Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"V__Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"V__Mood=Cnd|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Cnd|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Cnd|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Cnd|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Cnd|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Cnd|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Imp|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Imp|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Imp|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Imp|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Plur|Person=1|Tense=Past|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Plur|Person=2|Tense=Fut|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Plur|Person=3|Tense=Past|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Sing|Person=1|Tense=Fut|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Sing|Person=1|Tense=Past|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Sing|Person=2|Tense=Fut|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Sing|Person=3|Tense=Past|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Ind|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Sub|Number=Plur|Person=1|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Sub|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Sub|Number=Plur|Person=2|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Sub|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Sub|Number=Plur|Person=3|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Sub|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Sub|Number=Sing|Person=1|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Sub|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Sub|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Sub|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Sub|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin": {POS: VERB},
|
||||
"V__Mood=Sub|Number=Sing|Person=3|VerbForm=Fin": {POS: VERB},
|
||||
"V__Number=Plur|Tense=Pres|VerbForm=Part": {POS: VERB},
|
||||
"V__Number=Sing|Tense=Pres|VerbForm=Part": {POS: VERB},
|
||||
"V__Tense=Past|VerbForm=Part": {POS: VERB},
|
||||
"V__VerbForm=Ger": {POS: VERB},
|
||||
"V__VerbForm=Inf": {POS: VERB},
|
||||
"X___": {POS: X},
|
||||
"_SP": {POS: SPACE},
|
||||
}
|
||||
|
|
@ -1,7 +1,6 @@
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .punctuation import TOKENIZER_INFIXES
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .tag_map import TAG_MAP
|
||||
from .stop_words import STOP_WORDS
|
||||
|
||||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
|
|
@ -16,7 +15,6 @@ class LuxembourgishDefaults(Language.Defaults):
|
|||
lex_attr_getters[LANG] = lambda text: "lb"
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
stop_words = STOP_WORDS
|
||||
tag_map = TAG_MAP
|
||||
infixes = TOKENIZER_INFIXES
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -1,25 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, ADJ, CONJ, NUM, DET, ADV, ADP, X, VERB
|
||||
from ...symbols import NOUN, PART, SPACE, AUX
|
||||
|
||||
# TODO: tag map is still using POS tags from an internal training set.
|
||||
# These POS tags have to be modified to match those from Universal Dependencies
|
||||
|
||||
TAG_MAP = {
|
||||
"$": {POS: PUNCT},
|
||||
"ADJ": {POS: ADJ},
|
||||
"AV": {POS: ADV},
|
||||
"APPR": {POS: ADP, "AdpType": "prep"},
|
||||
"APPRART": {POS: ADP, "AdpType": "prep", "PronType": "art"},
|
||||
"D": {POS: DET, "PronType": "art"},
|
||||
"KO": {POS: CONJ},
|
||||
"N": {POS: NOUN},
|
||||
"P": {POS: ADV},
|
||||
"TRUNC": {POS: X, "Hyph": "yes"},
|
||||
"AUX": {POS: AUX},
|
||||
"V": {POS: VERB},
|
||||
"MV": {POS: VERB, "VerbType": "mod"},
|
||||
"PTK": {POS: PART},
|
||||
"INTER": {POS: PART},
|
||||
"NUM": {POS: NUM},
|
||||
"_SP": {POS: SPACE},
|
||||
}
|
||||
|
|
@ -2,8 +2,6 @@ from .punctuation import TOKENIZER_INFIXES, TOKENIZER_SUFFIXES
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .tag_map import TAG_MAP
|
||||
from .morph_rules import MORPH_RULES
|
||||
|
||||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
from ..norm_exceptions import BASE_NORMS
|
||||
|
|
@ -32,8 +30,6 @@ class LithuanianDefaults(Language.Defaults):
|
|||
del mod_base_exceptions["8)"]
|
||||
tokenizer_exceptions = update_exc(mod_base_exceptions, TOKENIZER_EXCEPTIONS)
|
||||
stop_words = STOP_WORDS
|
||||
tag_map = TAG_MAP
|
||||
morph_rules = MORPH_RULES
|
||||
|
||||
|
||||
class Lithuanian(Language):
|
||||
|
|
|
|||
File diff suppressed because it is too large
Load Diff
File diff suppressed because it is too large
Load Diff
|
|
@ -2,9 +2,7 @@ from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
|||
from .punctuation import TOKENIZER_PREFIXES, TOKENIZER_INFIXES
|
||||
from .punctuation import TOKENIZER_SUFFIXES
|
||||
from .stop_words import STOP_WORDS
|
||||
from .morph_rules import MORPH_RULES
|
||||
from .syntax_iterators import SYNTAX_ITERATORS
|
||||
from .tag_map import TAG_MAP
|
||||
|
||||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
from ..norm_exceptions import BASE_NORMS
|
||||
|
|
@ -24,8 +22,6 @@ class NorwegianDefaults(Language.Defaults):
|
|||
infixes = TOKENIZER_INFIXES
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
stop_words = STOP_WORDS
|
||||
morph_rules = MORPH_RULES
|
||||
tag_map = TAG_MAP
|
||||
syntax_iterators = SYNTAX_ITERATORS
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -1,665 +0,0 @@
|
|||
from ...symbols import LEMMA, PRON_LEMMA
|
||||
|
||||
# This dict includes all the PRON and DET tag combinations found in the
|
||||
# dataset developed by Schibsted, Nasjonalbiblioteket and LTG (to be published
|
||||
# autumn 2018) and the rarely used polite form.
|
||||
|
||||
MORPH_RULES = {
|
||||
"PRON__Animacy=Anim|Case=Nom|Number=Sing|Person=1|PronType=Prs": {
|
||||
"jeg": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Case": "Nom",
|
||||
}
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Number=Sing|Person=2|PronType=Prs": {
|
||||
"du": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Case": "Nom",
|
||||
},
|
||||
# polite form, not sure about the tag
|
||||
"De": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Case": "Nom",
|
||||
"Polite": "Form",
|
||||
},
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Gender=Fem|Number=Sing|Person=3|PronType=Prs": {
|
||||
"hun": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Fem",
|
||||
"Case": "Nom",
|
||||
}
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Gender=Masc|Number=Sing|Person=3|PronType=Prs": {
|
||||
"han": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
"Case": "Nom",
|
||||
}
|
||||
},
|
||||
"PRON__Gender=Neut|Number=Sing|Person=3|PronType=Prs": {
|
||||
"det": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
"alt": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
"intet": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
"noe": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Number": "Sing",
|
||||
"Person": "Three",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Number=Plur|Person=1|PronType=Prs": {
|
||||
"vi": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Case": "Nom",
|
||||
}
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Number=Plur|Person=2|PronType=Prs": {
|
||||
"dere": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Plur",
|
||||
"Case": "Nom",
|
||||
}
|
||||
},
|
||||
"PRON__Case=Nom|Number=Plur|Person=3|PronType=Prs": {
|
||||
"de": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Case": "Nom",
|
||||
}
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Acc|Number=Sing|Person=1|PronType=Prs": {
|
||||
"meg": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
}
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Acc|Number=Sing|Person=2|PronType=Prs": {
|
||||
"deg": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
},
|
||||
# polite form, not sure about the tag
|
||||
"Dem": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
"Polite": "Form",
|
||||
},
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Acc|Gender=Fem|Number=Sing|Person=3|PronType=Prs": {
|
||||
"henne": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Fem",
|
||||
"Case": "Acc",
|
||||
}
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Acc|Gender=Masc|Number=Sing|Person=3|PronType=Prs": {
|
||||
"ham": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"han": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
"Case": "Acc",
|
||||
},
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Acc|Number=Plur|Person=1|PronType=Prs": {
|
||||
"oss": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Case": "Acc",
|
||||
}
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Acc|Number=Plur|Person=2|PronType=Prs": {
|
||||
"dere": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Plur",
|
||||
"Case": "Acc",
|
||||
}
|
||||
},
|
||||
"PRON__Case=Acc|Number=Plur|Person=3|PronType=Prs": {
|
||||
"dem": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Case": "Acc",
|
||||
}
|
||||
},
|
||||
"PRON__Case=Acc|Reflex=Yes": {
|
||||
"seg": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"Person": "Three",
|
||||
"Number": "Sing,Plur",
|
||||
"Reflex": "Yes",
|
||||
}
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Number=Sing|PronType=Prs": {
|
||||
"man": {LEMMA: PRON_LEMMA, "PronType": "Prs", "Number": "Sing", "Case": "Nom"}
|
||||
},
|
||||
"DET__Gender=Masc|Number=Sing|Poss=Yes": {
|
||||
"min": {
|
||||
LEMMA: "min",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Masc",
|
||||
},
|
||||
"din": {
|
||||
LEMMA: "din",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Masc",
|
||||
},
|
||||
"hennes": {
|
||||
LEMMA: "hennes",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Masc",
|
||||
},
|
||||
"hans": {
|
||||
LEMMA: "hans",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Masc",
|
||||
},
|
||||
"sin": {
|
||||
LEMMA: "sin",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Masc",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"vår": {
|
||||
LEMMA: "vår",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Masc",
|
||||
},
|
||||
"deres": {
|
||||
LEMMA: "deres",
|
||||
"Person": "Two,Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Masc",
|
||||
},
|
||||
# polite form, not sure about the tag
|
||||
"Deres": {
|
||||
LEMMA: "Deres",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Masc",
|
||||
"Polite": "Form",
|
||||
},
|
||||
},
|
||||
"DET__Gender=Fem|Number=Sing|Poss=Yes": {
|
||||
"mi": {
|
||||
LEMMA: "min",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Fem",
|
||||
},
|
||||
"di": {
|
||||
LEMMA: "din",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Fem",
|
||||
},
|
||||
"hennes": {
|
||||
LEMMA: "hennes",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Fem",
|
||||
},
|
||||
"hans": {
|
||||
LEMMA: "hans",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Fem",
|
||||
},
|
||||
"si": {
|
||||
LEMMA: "sin",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Fem",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"vår": {
|
||||
LEMMA: "vår",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Fem",
|
||||
},
|
||||
"deres": {
|
||||
LEMMA: "deres",
|
||||
"Person": "Two,Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Fem",
|
||||
},
|
||||
# polite form, not sure about the tag
|
||||
"Deres": {
|
||||
LEMMA: "Deres",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Fem",
|
||||
"Polite": "Form",
|
||||
},
|
||||
},
|
||||
"DET__Gender=Neut|Number=Sing|Poss=Yes": {
|
||||
"mitt": {
|
||||
LEMMA: "min",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
"ditt": {
|
||||
LEMMA: "din",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
"hennes": {
|
||||
LEMMA: "hennes",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
"hans": {
|
||||
LEMMA: "hans",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
"sitt": {
|
||||
LEMMA: "sin",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Neut",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"vårt": {
|
||||
LEMMA: "vår",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
"deres": {
|
||||
LEMMA: "deres",
|
||||
"Person": "Two,Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
# polite form, not sure about the tag
|
||||
"Deres": {
|
||||
LEMMA: "Deres",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Gender": "Neut",
|
||||
"Polite": "Form",
|
||||
},
|
||||
},
|
||||
"DET__Number=Plur|Poss=Yes": {
|
||||
"mine": {LEMMA: "min", "Person": "One", "Number": "Plur", "Poss": "Yes"},
|
||||
"dine": {LEMMA: "din", "Person": "Two", "Number": "Plur", "Poss": "Yes"},
|
||||
"hennes": {LEMMA: "hennes", "Person": "Three", "Number": "Plur", "Poss": "Yes"},
|
||||
"hans": {LEMMA: "hans", "Person": "Three", "Number": "Plur", "Poss": "Yes"},
|
||||
"sine": {
|
||||
LEMMA: "sin",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"våre": {LEMMA: "vår", "Person": "One", "Number": "Plur", "Poss": "Yes"},
|
||||
"deres": {
|
||||
LEMMA: "deres",
|
||||
"Person": "Two,Three",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
},
|
||||
},
|
||||
"PRON__Animacy=Anim|Number=Plur|PronType=Rcp": {
|
||||
"hverandre": {LEMMA: PRON_LEMMA, "PronType": "Rcp", "Number": "Plur"}
|
||||
},
|
||||
"DET__Number=Plur|Poss=Yes|PronType=Rcp": {
|
||||
"hverandres": {
|
||||
LEMMA: "hverandres",
|
||||
"PronType": "Rcp",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
}
|
||||
},
|
||||
"PRON___": {"som": {LEMMA: PRON_LEMMA}, "ikkenoe": {LEMMA: PRON_LEMMA}},
|
||||
"PRON__PronType=Int": {"hva": {LEMMA: PRON_LEMMA, "PronType": "Int"}},
|
||||
"PRON__Animacy=Anim|PronType=Int": {"hvem": {LEMMA: PRON_LEMMA, "PronType": "Int"}},
|
||||
"PRON__Animacy=Anim|Poss=Yes|PronType=Int": {
|
||||
"hvis": {LEMMA: PRON_LEMMA, "PronType": "Int", "Poss": "Yes"}
|
||||
},
|
||||
"PRON__Number=Plur|Person=3|PronType=Prs": {
|
||||
"noen": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Number": "Plur",
|
||||
"Person": "Three",
|
||||
},
|
||||
"ingen": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Number": "Plur",
|
||||
"Person": "Three",
|
||||
},
|
||||
"alle": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Number": "Plur",
|
||||
"Person": "Three",
|
||||
},
|
||||
},
|
||||
"PRON__Gender=Fem,Masc|Number=Sing|Person=3|PronType=Prs": {
|
||||
"noen": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Number": "Sing",
|
||||
"Person": "Three",
|
||||
"Gender": "Fem,Masc",
|
||||
},
|
||||
"den": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Number": "Sing",
|
||||
"Person": "Three",
|
||||
"Gender": "Fem,Masc",
|
||||
},
|
||||
"ingen": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Number": "Sing",
|
||||
"Person": "Three",
|
||||
"Gender": "Fem,Masc",
|
||||
"Polarity": "Neg",
|
||||
},
|
||||
},
|
||||
"PRON__Number=Sing": {"ingenting": {LEMMA: PRON_LEMMA, "Number": "Sing"}},
|
||||
"PRON__Animacy=Anim|Number=Sing|PronType=Prs": {
|
||||
"en": {LEMMA: PRON_LEMMA, "PronType": "Prs", "Number": "Sing"}
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Gen,Nom|Number=Sing|PronType=Prs": {
|
||||
"ens": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Number": "Sing",
|
||||
"Case": "Gen,Nom",
|
||||
}
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Gen|Number=Sing|PronType=Prs": {
|
||||
"ens": {LEMMA: PRON_LEMMA, "PronType": "Prs", "Number": "Sing", "Case": "Gen"}
|
||||
},
|
||||
"DET__Case=Gen|Gender=Masc|Number=Sing": {
|
||||
"ens": {LEMMA: "en", "Number": "Sing", "Case": "Gen"}
|
||||
},
|
||||
"DET__Gender=Masc|Number=Sing": {
|
||||
"enhver": {LEMMA: "enhver", "Number": "Sing", "Gender": "Masc"},
|
||||
"all": {LEMMA: "all", "Number": "Sing", "Gender": "Masc"},
|
||||
"hver": {LEMMA: "hver", "Number": "Sing", "Gender": "Masc"},
|
||||
"noen": {LEMMA: "noen", "Gender": "Masc", "Number": "Sing"},
|
||||
"noe": {LEMMA: "noen", "Gender": "Masc", "Number": "Sing"},
|
||||
"en": {LEMMA: "en", "Number": "Sing", "Gender": "Neut"},
|
||||
"ingen": {LEMMA: "ingen", "Gender": "Masc", "Number": "Sing"},
|
||||
},
|
||||
"DET__Gender=Fem|Number=Sing": {
|
||||
"enhver": {LEMMA: "enhver", "Number": "Sing", "Gender": "Fem"},
|
||||
"all": {LEMMA: "all", "Number": "Sing", "Gender": "Fem"},
|
||||
"hver": {LEMMA: "hver", "Number": "Sing", "Gender": "Fem"},
|
||||
"noen": {LEMMA: "noen", "Gender": "Fem", "Number": "Sing"},
|
||||
"noe": {LEMMA: "noen", "Gender": "Fem", "Number": "Sing"},
|
||||
"ei": {LEMMA: "en", "Number": "Sing", "Gender": "Fem"},
|
||||
},
|
||||
"DET__Gender=Neut|Number=Sing": {
|
||||
"ethvert": {LEMMA: "enhver", "Number": "Sing", "Gender": "Neut"},
|
||||
"alt": {LEMMA: "all", "Number": "Sing", "Gender": "Neut"},
|
||||
"hvert": {LEMMA: "hver", "Number": "Sing", "Gender": "Neut"},
|
||||
"noe": {LEMMA: "noen", "Number": "Sing", "Gender": "Neut"},
|
||||
"intet": {LEMMA: "ingen", "Gender": "Neut", "Number": "Sing"},
|
||||
"et": {LEMMA: "en", "Number": "Sing", "Gender": "Neut"},
|
||||
},
|
||||
"DET__Gender=Neut|Number=Sing|PronType=Int": {
|
||||
"hvilket": {
|
||||
LEMMA: "hvilken",
|
||||
"PronType": "Int",
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
}
|
||||
},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Int": {
|
||||
"hvilken": {
|
||||
LEMMA: "hvilken",
|
||||
"PronType": "Int",
|
||||
"Number": "Sing",
|
||||
"Gender": "Fem",
|
||||
}
|
||||
},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Int": {
|
||||
"hvilken": {
|
||||
LEMMA: "hvilken",
|
||||
"PronType": "Int",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
}
|
||||
},
|
||||
"DET__Number=Plur|PronType=Int": {
|
||||
"hvilke": {LEMMA: "hvilken", "PronType": "Int", "Number": "Plur"}
|
||||
},
|
||||
"DET__Number=Plur": {
|
||||
"alle": {LEMMA: "all", "Number": "Plur"},
|
||||
"noen": {LEMMA: "noen", "Number": "Plur"},
|
||||
"egne": {LEMMA: "egen", "Number": "Plur"},
|
||||
"ingen": {LEMMA: "ingen", "Number": "Plur"},
|
||||
},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Dem": {
|
||||
"den": {LEMMA: "den", "PronType": "Dem", "Number": "Sing", "Gender": "Masc"},
|
||||
"slik": {LEMMA: "slik", "PronType": "Dem", "Number": "Sing", "Gender": "Masc"},
|
||||
"denne": {
|
||||
LEMMA: "denne",
|
||||
"PronType": "Dem",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
},
|
||||
},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Dem": {
|
||||
"den": {LEMMA: "den", "PronType": "Dem", "Number": "Sing", "Gender": "Fem"},
|
||||
"slik": {LEMMA: "slik", "PronType": "Dem", "Number": "Sing", "Gender": "Fem"},
|
||||
"denne": {LEMMA: "denne", "PronType": "Dem", "Number": "Sing", "Gender": "Fem"},
|
||||
},
|
||||
"DET__Gender=Neut|Number=Sing|PronType=Dem": {
|
||||
"det": {LEMMA: "det", "PronType": "Dem", "Number": "Sing", "Gender": "Neut"},
|
||||
"slikt": {LEMMA: "slik", "PronType": "Dem", "Number": "Sing", "Gender": "Neut"},
|
||||
"dette": {
|
||||
LEMMA: "dette",
|
||||
"PronType": "Dem",
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
},
|
||||
"DET__Number=Plur|PronType=Dem": {
|
||||
"disse": {LEMMA: "disse", "PronType": "Dem", "Number": "Plur"},
|
||||
"andre": {LEMMA: "annen", "PronType": "Dem", "Number": "Plur"},
|
||||
"de": {LEMMA: "de", "PronType": "Dem", "Number": "Plur"},
|
||||
"slike": {LEMMA: "slik", "PronType": "Dem", "Number": "Plur"},
|
||||
},
|
||||
"DET__Definite=Ind|Gender=Masc|Number=Sing|PronType=Dem": {
|
||||
"annen": {LEMMA: "annen", "PronType": "Dem", "Number": "Sing", "Gender": "Masc"}
|
||||
},
|
||||
"DET__Definite=Ind|Gender=Fem|Number=Sing|PronType=Dem": {
|
||||
"annen": {LEMMA: "annen", "PronType": "Dem", "Number": "Sing", "Gender": "Fem"}
|
||||
},
|
||||
"DET__Definite=Ind|Gender=Neut|Number=Sing|PronType=Dem": {
|
||||
"annet": {LEMMA: "annen", "PronType": "Dem", "Number": "Sing", "Gender": "Neut"}
|
||||
},
|
||||
"DET__Case=Gen|Definite=Ind|Gender=Masc|Number=Sing|PronType=Dem": {
|
||||
"annens": {
|
||||
LEMMA: "annnen",
|
||||
"PronType": "Dem",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
"Case": "Gen",
|
||||
}
|
||||
},
|
||||
"DET__Case=Gen|Number=Plur|PronType=Dem": {
|
||||
"andres": {LEMMA: "annen", "PronType": "Dem", "Number": "Plur", "Case": "Gen"}
|
||||
},
|
||||
"DET__Case=Gen|Gender=Fem|Number=Sing|PronType=Dem": {
|
||||
"dens": {
|
||||
LEMMA: "den",
|
||||
"PronType": "Dem",
|
||||
"Number": "Sing",
|
||||
"Gender": "Fem",
|
||||
"Case": "Gen",
|
||||
}
|
||||
},
|
||||
"DET__Case=Gen|Gender=Masc|Number=Sing|PronType=Dem": {
|
||||
"hvis": {
|
||||
LEMMA: "hvis",
|
||||
"PronType": "Dem",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
"Case": "Gen",
|
||||
},
|
||||
"dens": {
|
||||
LEMMA: "den",
|
||||
"PronType": "Dem",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
"Case": "Gen",
|
||||
},
|
||||
},
|
||||
"DET__Case=Gen|Gender=Neut|Number=Sing|PronType=Dem": {
|
||||
"dets": {
|
||||
LEMMA: "det",
|
||||
"PronType": "Dem",
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
"Case": "Gen",
|
||||
}
|
||||
},
|
||||
"DET__Case=Gen|Number=Plur": {
|
||||
"alles": {LEMMA: "all", "Number": "Plur", "Case": "Gen"}
|
||||
},
|
||||
"DET__Definite=Def|Number=Sing|PronType=Dem": {
|
||||
"andre": {LEMMA: "annen", "Number": "Sing", "PronType": "Dem"}
|
||||
},
|
||||
"DET__Definite=Def|PronType=Dem": {
|
||||
"samme": {LEMMA: "samme", "PronType": "Dem"},
|
||||
"forrige": {LEMMA: "forrige", "PronType": "Dem"},
|
||||
"neste": {LEMMA: "neste", "PronType": "Dem"},
|
||||
},
|
||||
"DET__Definite=Def": {"selve": {LEMMA: "selve"}, "selveste": {LEMMA: "selveste"}},
|
||||
"DET___": {"selv": {LEMMA: "selv"}, "endel": {LEMMA: "endel"}},
|
||||
"DET__Definite=Ind|Gender=Fem|Number=Sing": {
|
||||
"egen": {LEMMA: "egen", "Gender": "Fem", "Number": "Sing"}
|
||||
},
|
||||
"DET__Definite=Ind|Gender=Masc|Number=Sing": {
|
||||
"egen": {LEMMA: "egen", "Gender": "Masc", "Number": "Sing"}
|
||||
},
|
||||
"DET__Definite=Ind|Gender=Neut|Number=Sing": {
|
||||
"eget": {LEMMA: "egen", "Gender": "Neut", "Number": "Sing"}
|
||||
},
|
||||
# same wordform and pos (verb), have to specify the exact features in order to not mix them up
|
||||
"VERB__Mood=Ind|Tense=Pres|VerbForm=Fin": {
|
||||
"så": {LEMMA: "så", "VerbForm": "Fin", "Tense": "Pres", "Mood": "Ind"}
|
||||
},
|
||||
"VERB__Mood=Ind|Tense=Past|VerbForm=Fin": {
|
||||
"så": {LEMMA: "se", "VerbForm": "Fin", "Tense": "Past", "Mood": "Ind"}
|
||||
},
|
||||
}
|
||||
|
||||
# copied from the English morph_rules.py
|
||||
for tag, rules in MORPH_RULES.items():
|
||||
for key, attrs in dict(rules).items():
|
||||
rules[key.title()] = attrs
|
||||
|
|
@ -1,758 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, ADJ, CONJ, CCONJ, SCONJ, SYM, NUM, DET, ADV, ADP, X
|
||||
from ...symbols import VERB, NOUN, PROPN, PART, INTJ, PRON, AUX
|
||||
|
||||
|
||||
# Tags are a combination of POS and morphological features from a
|
||||
# https://github.com/ltgoslo/norne developed by Schibsted, Nasjonalbiblioteket and LTG. The
|
||||
# data format is .conllu and follows the Universal Dependencies annotation.
|
||||
# (There are some annotation differences compared to this dataset:
|
||||
# https://github.com/UniversalDependencies/UD_Norwegian-Bokmaal
|
||||
# mainly in the way determiners and pronouns are tagged).
|
||||
TAG_MAP = {
|
||||
"ADJ__Case=Gen|Definite=Def|Degree=Pos|Number=Sing": {
|
||||
"morph": "Case=Gen|Definite=Def|Degree=Pos|Number=Sing",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Case=Gen|Definite=Def|Number=Sing": {
|
||||
"morph": "Case=Gen|Definite=Def|Number=Sing",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Case=Gen|Definite=Ind|Degree=Pos|Gender=Neut|Number=Sing": {
|
||||
"morph": "Case=Gen|Definite=Ind|Degree=Pos|Gender=Neut|Number=Sing",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Case=Gen|Definite=Ind|Degree=Pos|Number=Sing": {
|
||||
"morph": "Case=Gen|Definite=Ind|Degree=Pos|Number=Sing",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Case=Gen|Degree=Cmp": {"morph": "Case=Gen|Degree=Cmp", POS: ADJ},
|
||||
"ADJ__Case=Gen|Degree=Pos|Number=Plur": {
|
||||
"morph": "Case=Gen|Degree=Pos|Number=Plur",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Definite=Def|Degree=Pos|Gender=Masc|Number=Sing": {
|
||||
"morph": "Definite=Def|Degree=Pos|Gender=Masc|Number=Sing",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Definite=Def|Degree=Pos|Number=Sing": {
|
||||
"morph": "Definite=Def|Degree=Pos|Number=Sing",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Definite=Def|Degree=Sup": {"morph": "Definite=Def|Degree=Sup", POS: ADJ},
|
||||
"ADJ__Definite=Def|Number=Sing": {"morph": "Definite=Def|Number=Sing", POS: ADJ},
|
||||
"ADJ__Definite=Ind|Degree=Pos": {"morph": "Definite=Ind|Degree=Pos", POS: ADJ},
|
||||
"ADJ__Definite=Ind|Degree=Pos|Gender=Masc|Number=Sing": {
|
||||
"morph": "Definite=Ind|Degree=Pos|Gender=Masc|Number=Sing",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Definite=Ind|Degree=Pos|Gender=Neut|Number=Sing": {
|
||||
"morph": "Definite=Ind|Degree=Pos|Gender=Neut|Number=Sing",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Definite=Ind|Degree=Pos|Number=Sing": {
|
||||
"morph": "Definite=Ind|Degree=Pos|Number=Sing",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Definite=Ind|Degree=Sup": {"morph": "Definite=Ind|Degree=Sup", POS: ADJ},
|
||||
"ADJ__Definite=Ind|Gender=Masc|Number=Sing": {
|
||||
"morph": "Definite=Ind|Gender=Masc|Number=Sing",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Definite=Ind|Gender=Neut|Number=Sing": {
|
||||
"morph": "Definite=Ind|Gender=Neut|Number=Sing",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Definite=Ind|Number=Sing": {"morph": "Definite=Ind|Number=Sing", POS: ADJ},
|
||||
"ADJ__Degree=Cmp": {"morph": "Degree=Cmp", POS: ADJ},
|
||||
"ADJ__Degree=Pos": {"morph": "Degree=Pos", POS: ADJ},
|
||||
"ADJ__Degree=Pos|Number=Plur": {"morph": "Degree=Pos|Number=Plur", POS: ADJ},
|
||||
"ADJ__Degree=Sup": {"morph": "Degree=Sup", POS: ADJ},
|
||||
"ADJ__Number=Plur": {"morph": "Number=Plur", POS: ADJ},
|
||||
"ADJ__Number=Plur|VerbForm=Part": {"morph": "Number=Plur|VerbForm=Part", POS: ADJ},
|
||||
"ADJ__Number=Sing": {"morph": "Number=Sing", POS: ADJ},
|
||||
"ADJ___": {"morph": "_", POS: ADJ},
|
||||
"ADP___": {"morph": "_", POS: ADP},
|
||||
"ADV___": {"morph": "_", POS: ADV},
|
||||
"ADV__Gender=Masc": {"morph": "Gender=Masc", POS: ADV},
|
||||
"AUX__Mood=Imp|VerbForm=Fin": {"morph": "Mood=Imp|VerbForm=Fin", POS: AUX},
|
||||
"AUX__Mood=Ind|Tense=Past|VerbForm=Fin": {
|
||||
"morph": "Mood=Ind|Tense=Past|VerbForm=Fin",
|
||||
POS: AUX,
|
||||
},
|
||||
"AUX__Mood=Ind|Tense=Pres|VerbForm=Fin": {
|
||||
"morph": "Mood=Ind|Tense=Pres|VerbForm=Fin",
|
||||
POS: AUX,
|
||||
},
|
||||
"AUX__Mood=Ind|Tense=Pres|VerbForm=Fin|Voice=Pass": {
|
||||
"morph": "Mood=Ind|Tense=Pres|VerbForm=Fin|Voice=Pass",
|
||||
POS: AUX,
|
||||
},
|
||||
"AUX__VerbForm=Inf": {"morph": "VerbForm=Inf", POS: AUX},
|
||||
"AUX__VerbForm=Inf|Voice=Pass": {"morph": "VerbForm=Inf|Voice=Pass", POS: AUX},
|
||||
"AUX__VerbForm=Part": {"morph": "VerbForm=Part", POS: AUX},
|
||||
"CONJ___": {"morph": "_", POS: CONJ},
|
||||
"DET__Case=Gen|Definite=Ind|Gender=Masc|Number=Sing|PronType=Dem": {
|
||||
"morph": "Case=Gen|Definite=Ind|Gender=Masc|Number=Sing|PronType=Dem",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Case=Gen|Gender=Fem|Number=Sing|PronType=Dem": {
|
||||
"morph": "Case=Gen|Gender=Fem|Number=Sing|PronType=Dem",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Case=Gen|Gender=Masc|Number=Sing": {
|
||||
"morph": "Case=Gen|Gender=Masc|Number=Sing",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Case=Gen|Gender=Masc|Number=Sing|PronType=Dem": {
|
||||
"morph": "Case=Gen|Gender=Masc|Number=Sing|PronType=Dem",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Case=Gen|Gender=Neut|Number=Sing|PronType=Dem": {
|
||||
"morph": "Case=Gen|Gender=Neut|Number=Sing|PronType=Dem",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Case=Gen|Number=Plur": {"morph": "Case=Gen|Number=Plur", POS: DET},
|
||||
"DET__Case=Gen|Number=Plur|PronType=Dem": {
|
||||
"morph": "Case=Gen|Number=Plur|PronType=Dem",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Definite=Def": {"morph": "Definite=Def", POS: DET},
|
||||
"DET__Definite=Def|Number=Sing|PronType=Dem": {
|
||||
"morph": "Definite=Def|Number=Sing|PronType=Dem",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Definite=Def|PronType=Dem": {"morph": "Definite=Def|PronType=Dem", POS: DET},
|
||||
"DET__Definite=Ind|Gender=Fem|Number=Sing": {
|
||||
"morph": "Definite=Ind|Gender=Fem|Number=Sing",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Definite=Ind|Gender=Fem|Number=Sing|PronType=Dem": {
|
||||
"morph": "Definite=Ind|Gender=Fem|Number=Sing|PronType=Dem",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Definite=Ind|Gender=Masc|Number=Sing": {
|
||||
"morph": "Definite=Ind|Gender=Masc|Number=Sing",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Definite=Ind|Gender=Masc|Number=Sing|PronType=Dem": {
|
||||
"morph": "Definite=Ind|Gender=Masc|Number=Sing|PronType=Dem",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Definite=Ind|Gender=Neut|Number=Sing": {
|
||||
"morph": "Definite=Ind|Gender=Neut|Number=Sing",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Definite=Ind|Gender=Neut|Number=Sing|PronType=Dem": {
|
||||
"morph": "Definite=Ind|Gender=Neut|Number=Sing|PronType=Dem",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Degree=Pos|Number=Plur": {"morph": "Degree=Pos|Number=Plur", POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing": {"morph": "Gender=Fem|Number=Sing", POS: DET},
|
||||
"DET__Gender=Fem|Number=Sing|Poss=Yes": {
|
||||
"morph": "Gender=Fem|Number=Sing|Poss=Yes",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Dem": {
|
||||
"morph": "Gender=Fem|Number=Sing|PronType=Dem",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Int": {
|
||||
"morph": "Gender=Fem|Number=Sing|PronType=Int",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Masc|Number=Sing": {"morph": "Gender=Masc|Number=Sing", POS: DET},
|
||||
"DET__Gender=Masc|Number=Sing|Poss=Yes": {
|
||||
"morph": "Gender=Masc|Number=Sing|Poss=Yes",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Dem": {
|
||||
"morph": "Gender=Masc|Number=Sing|PronType=Dem",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Int": {
|
||||
"morph": "Gender=Masc|Number=Sing|PronType=Int",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Neut|Number=Sing": {"morph": "Gender=Neut|Number=Sing", POS: DET},
|
||||
"DET__Gender=Neut|Number=Sing|Poss=Yes": {
|
||||
"morph": "Gender=Neut|Number=Sing|Poss=Yes",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Neut|Number=Sing|PronType=Dem": {
|
||||
"morph": "Gender=Neut|Number=Sing|PronType=Dem",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Neut|Number=Sing|PronType=Int": {
|
||||
"morph": "Gender=Neut|Number=Sing|PronType=Int",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Number=Plur": {"morph": "Number=Plur", POS: DET},
|
||||
"DET__Number=Plur|Poss=Yes": {"morph": "Number=Plur|Poss=Yes", POS: DET},
|
||||
"DET__Number=Plur|Poss=Yes|PronType=Rcp": {
|
||||
"morph": "Number=Plur|Poss=Yes|PronType=Rcp",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Number=Plur|PronType=Dem": {"morph": "Number=Plur|PronType=Dem", POS: DET},
|
||||
"DET__Number=Plur|PronType=Int": {"morph": "Number=Plur|PronType=Int", POS: DET},
|
||||
"DET___": {"morph": "_", POS: DET},
|
||||
"INTJ___": {"morph": "_", POS: INTJ},
|
||||
"NOUN__Case=Gen": {"morph": "Case=Gen", POS: NOUN},
|
||||
"NOUN__Case=Gen|Definite=Def|Gender=Fem|Number=Plur": {
|
||||
"morph": "Case=Gen|Definite=Def|Gender=Fem|Number=Plur",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Case=Gen|Definite=Def|Gender=Fem|Number=Sing": {
|
||||
"morph": "Case=Gen|Definite=Def|Gender=Fem|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Case=Gen|Definite=Def|Gender=Masc|Number=Plur": {
|
||||
"morph": "Case=Gen|Definite=Def|Gender=Masc|Number=Plur",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Case=Gen|Definite=Def|Gender=Masc|Number=Sing": {
|
||||
"morph": "Case=Gen|Definite=Def|Gender=Masc|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Case=Gen|Definite=Def|Gender=Neut|Number=Plur": {
|
||||
"morph": "Case=Gen|Definite=Def|Gender=Neut|Number=Plur",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Case=Gen|Definite=Def|Gender=Neut|Number=Sing": {
|
||||
"morph": "Case=Gen|Definite=Def|Gender=Neut|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Case=Gen|Definite=Ind|Gender=Fem|Number=Plur": {
|
||||
"morph": "Case=Gen|Definite=Ind|Gender=Fem|Number=Plur",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Case=Gen|Definite=Ind|Gender=Fem|Number=Sing": {
|
||||
"morph": "Case=Gen|Definite=Ind|Gender=Fem|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Case=Gen|Definite=Ind|Gender=Masc|Number=Plur": {
|
||||
"morph": "Case=Gen|Definite=Ind|Gender=Masc|Number=Plur",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Case=Gen|Definite=Ind|Gender=Masc|Number=Sing": {
|
||||
"morph": "Case=Gen|Definite=Ind|Gender=Masc|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Case=Gen|Definite=Ind|Gender=Neut|Number=Plur": {
|
||||
"morph": "Case=Gen|Definite=Ind|Gender=Neut|Number=Plur",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Case=Gen|Definite=Ind|Gender=Neut|Number=Sing": {
|
||||
"morph": "Case=Gen|Definite=Ind|Gender=Neut|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Case=Gen|Gender=Fem": {"morph": "Case=Gen|Gender=Fem", POS: NOUN},
|
||||
"NOUN__Definite=Def,Ind|Gender=Masc|Number=Plur,Sing": {
|
||||
"morph": "Definite=Def",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Def,Ind|Gender=Masc|Number=Sing": {
|
||||
"morph": "Definite=Def",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Def,Ind|Gender=Neut|Number=Plur,Sing": {
|
||||
"morph": "Definite=Def",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Def|Gender=Fem|Number=Plur": {
|
||||
"morph": "Definite=Def|Gender=Fem|Number=Plur",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Def|Gender=Fem|Number=Sing": {
|
||||
"morph": "Definite=Def|Gender=Fem|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Def|Gender=Masc|Number=Plur": {
|
||||
"morph": "Definite=Def|Gender=Masc|Number=Plur",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Def|Gender=Masc|Number=Sing": {
|
||||
"morph": "Definite=Def|Gender=Masc|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Def|Gender=Neut|Number=Plur": {
|
||||
"morph": "Definite=Def|Gender=Neut|Number=Plur",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Def|Gender=Neut|Number=Sing": {
|
||||
"morph": "Definite=Def|Gender=Neut|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Def|Number=Plur": {"morph": "Definite=Def|Number=Plur", POS: NOUN},
|
||||
"NOUN__Definite=Ind|Gender=Fem|Number=Plur": {
|
||||
"morph": "Definite=Ind|Gender=Fem|Number=Plur",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Ind|Gender=Fem|Number=Sing": {
|
||||
"morph": "Definite=Ind|Gender=Fem|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Ind|Gender=Masc": {"morph": "Definite=Ind|Gender=Masc", POS: NOUN},
|
||||
"NOUN__Definite=Ind|Gender=Masc|Number=Plur": {
|
||||
"morph": "Definite=Ind|Gender=Masc|Number=Plur",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Ind|Gender=Masc|Number=Sing": {
|
||||
"morph": "Definite=Ind|Gender=Masc|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Ind|Gender=Neut|Number=Plur": {
|
||||
"morph": "Definite=Ind|Gender=Neut|Number=Plur",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Ind|Gender=Neut|Number=Sing": {
|
||||
"morph": "Definite=Ind|Gender=Neut|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Definite=Ind|Number=Plur": {"morph": "Definite=Ind|Number=Plur", POS: NOUN},
|
||||
"NOUN__Definite=Ind|Number=Sing": {"morph": "Definite=Ind|Number=Sing", POS: NOUN},
|
||||
"NOUN__Gender=Fem": {"morph": "Gender=Fem", POS: NOUN},
|
||||
"NOUN__Gender=Masc": {"morph": "Gender=Masc", POS: NOUN},
|
||||
"NOUN__Gender=Masc|Number=Sing": {"morph": "Gender=Masc|Number=Sing", POS: NOUN},
|
||||
"NOUN__Gender=Neut": {"morph": "Gender=Neut", POS: NOUN},
|
||||
"NOUN__Number=Plur": {"morph": "Number=Plur", POS: NOUN},
|
||||
"NOUN___": {"morph": "_", POS: NOUN},
|
||||
"NUM__Case=Gen|Number=Plur": {"morph": "Case=Gen|Number=Plur", POS: NUM},
|
||||
"NUM__Definite=Def": {"morph": "Definite=Def", POS: NUM},
|
||||
"NUM__Definite=Def|Number=Sing": {"morph": "Definite=Def|Number=Sing", POS: NUM},
|
||||
"NUM__Gender=Fem|Number=Sing": {"morph": "Gender=Fem|Number=Sing", POS: NUM},
|
||||
"NUM__Gender=Masc|Number=Sing": {"morph": "Gender=Masc|Number=Sing", POS: NUM},
|
||||
"NUM__Gender=Neut|Number=Sing": {"morph": "Gender=Neut|Number=Sing", POS: NUM},
|
||||
"NUM__Number=Plur": {"morph": "Number=Plur", POS: NUM},
|
||||
"NUM__Number=Sing": {"morph": "Number=Sing", POS: NUM},
|
||||
"NUM___": {"morph": "_", POS: NUM},
|
||||
"PART___": {"morph": "_", POS: PART},
|
||||
"PRON__Animacy=Anim|Case=Acc|Gender=Fem|Number=Sing|Person=3|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Acc|Gender=Fem|Number=Sing|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Acc|Gender=Masc|Number=Sing|Person=3|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Acc|Gender=Masc|Number=Sing|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Acc|Number=Plur|Person=1|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Acc|Number=Plur|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Acc|Number=Plur|Person=2|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Acc|Number=Plur|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Acc|Number=Sing|Person=1|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Acc|Number=Sing|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Acc|Number=Sing|Person=2|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Acc|Number=Sing|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Gen,Nom|Number=Sing|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Gen",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Gen|Number=Sing|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Gen|Number=Sing|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Gender=Fem|Number=Sing|Person=3|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Nom|Gender=Fem|Number=Sing|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Gender=Masc|Number=Sing|Person=3|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Nom|Gender=Masc|Number=Sing|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Number=Plur|Person=1|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Nom|Number=Plur|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Number=Plur|Person=2|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Nom|Number=Plur|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Number=Sing|Person=1|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Nom|Number=Sing|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Number=Sing|Person=2|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Nom|Number=Sing|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Case=Nom|Number=Sing|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Case=Nom|Number=Sing|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Number=Plur|PronType=Rcp": {
|
||||
"morph": "Animacy=Anim|Number=Plur|PronType=Rcp",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Number=Sing|PronType=Prs": {
|
||||
"morph": "Animacy=Anim|Number=Sing|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|Poss=Yes|PronType=Int": {
|
||||
"morph": "Animacy=Anim|Poss=Yes|PronType=Int",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Anim|PronType=Int": {
|
||||
"morph": "Animacy=Anim|PronType=Int",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Case=Acc|Number=Plur|Person=3|PronType=Prs": {
|
||||
"morph": "Case=Acc|Number=Plur|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Case=Acc|Reflex=Yes": {"morph": "Case=Acc|Reflex=Yes", POS: PRON},
|
||||
"PRON__Case=Nom|Number=Plur|Person=3|PronType=Prs": {
|
||||
"morph": "Case=Nom|Number=Plur|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Case=Gen|Number=Plur|Person=3|PronType=Prs": {
|
||||
"morph": "Case=Gen|Number=Plur|Person=3|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Gender=Fem,Masc|Number=Sing|Person=3|PronType=Prs": {
|
||||
"morph": "Gender=Fem",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Gender=Neut|Number=Sing|Person=3|PronType=Prs": {
|
||||
"morph": "Gender=Neut|Number=Sing|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Number=Plur|Person=3|PronType=Prs": {
|
||||
"morph": "Number=Plur|Person=",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Number=Sing": {"morph": "Number=Sing", POS: PRON},
|
||||
"PRON__PronType=Int": {"morph": "PronType=Int", POS: PRON},
|
||||
"PRON___": {"morph": "_", POS: PRON},
|
||||
"PROPN__Case=Gen": {"morph": "Case=Gen", POS: PROPN},
|
||||
"PROPN__Case=Gen|Gender=Fem": {"morph": "Case=Gen|Gender=Fem", POS: PROPN},
|
||||
"PROPN__Case=Gen|Gender=Masc": {"morph": "Case=Gen|Gender=Masc", POS: PROPN},
|
||||
"PROPN__Case=Gen|Gender=Neut": {"morph": "Case=Gen|Gender=Neut", POS: PROPN},
|
||||
"PROPN__Gender=Fem": {"morph": "Gender=Fem", POS: PROPN},
|
||||
"PROPN__Gender=Masc": {"morph": "Gender=Masc", POS: PROPN},
|
||||
"PROPN__Gender=Neut": {"morph": "Gender=Neut", POS: PROPN},
|
||||
"PROPN___": {"morph": "_", POS: PROPN},
|
||||
"PUNCT___": {"morph": "_", POS: PUNCT},
|
||||
"SCONJ___": {"morph": "_", POS: SCONJ},
|
||||
"SYM___": {"morph": "_", POS: SYM},
|
||||
"VERB__Definite=Ind|Number=Sing": {"morph": "Definite=Ind|Number=Sing", POS: VERB},
|
||||
"VERB__Mood=Imp|VerbForm=Fin": {"morph": "Mood=Imp|VerbForm=Fin", POS: VERB},
|
||||
"VERB__Mood=Ind|Tense=Past|VerbForm=Fin": {
|
||||
"morph": "Mood=Ind|Tense=Past|VerbForm=Fin",
|
||||
POS: VERB,
|
||||
},
|
||||
"VERB__Mood=Ind|Tense=Past|VerbForm=Fin|Voice=Pass": {
|
||||
"morph": "Mood=Ind|Tense=Past|VerbForm=Fin|Voice=Pass",
|
||||
POS: VERB,
|
||||
},
|
||||
"VERB__Mood=Ind|Tense=Pres|VerbForm=Fin": {
|
||||
"morph": "Mood=Ind|Tense=Pres|VerbForm=Fin",
|
||||
POS: VERB,
|
||||
},
|
||||
"VERB__Mood=Ind|Tense=Pres|VerbForm=Fin|Voice=Pass": {
|
||||
"morph": "Mood=Ind|Tense=Pres|VerbForm=Fin|Voice=Pass",
|
||||
POS: VERB,
|
||||
},
|
||||
"VERB__VerbForm=Inf": {"morph": "VerbForm=Inf", POS: VERB},
|
||||
"VERB__VerbForm=Inf|Voice=Pass": {"morph": "VerbForm=Inf|Voice=Pass", POS: VERB},
|
||||
"VERB__VerbForm=Part": {"morph": "VerbForm=Part", POS: VERB},
|
||||
"VERB___": {"morph": "_", POS: VERB},
|
||||
"X___": {"morph": "_", POS: X},
|
||||
"CCONJ___": {"morph": "_", POS: CCONJ},
|
||||
"ADJ__Abbr=Yes": {"morph": "Abbr=Yes", POS: ADJ},
|
||||
"ADJ__Abbr=Yes|Degree=Pos": {"morph": "Abbr=Yes|Degree=Pos", POS: ADJ},
|
||||
"ADJ__Case=Gen|Definite=Def|Number=Sing|VerbForm=Part": {
|
||||
"morph": "Case=Gen|Definite=Def|Number=Sing|VerbForm=Part",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Definite=Def|Number=Sing|VerbForm=Part": {
|
||||
"morph": "Definite=Def|Number=Sing|VerbForm=Part",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Definite=Ind|Gender=Masc|Number=Sing|VerbForm=Part": {
|
||||
"morph": "Definite=Ind|Gender=Masc|Number=Sing|VerbForm=Part",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Definite=Ind|Gender=Neut|Number=Sing|VerbForm=Part": {
|
||||
"morph": "Definite=Ind|Gender=Neut|Number=Sing|VerbForm=Part",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Definite=Ind|Number=Sing|VerbForm=Part": {
|
||||
"morph": "Definite=Ind|Number=Sing|VerbForm=Part",
|
||||
POS: ADJ,
|
||||
},
|
||||
"ADJ__Number=Sing|VerbForm=Part": {"morph": "Number=Sing|VerbForm=Part", POS: ADJ},
|
||||
"ADJ__VerbForm=Part": {"morph": "VerbForm=Part", POS: ADJ},
|
||||
"ADP__Abbr=Yes": {"morph": "Abbr=Yes", POS: ADP},
|
||||
"ADV__Abbr=Yes": {"morph": "Abbr=Yes", POS: ADV},
|
||||
"DET__Case=Gen|Gender=Masc|Number=Sing|PronType=Art": {
|
||||
"morph": "Case=Gen|Gender=Masc|Number=Sing|PronType=Art",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Case=Gen|Number=Plur|PronType=Tot": {
|
||||
"morph": "Case=Gen|Number=Plur|PronType=Tot",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Definite=Def|PronType=Prs": {"morph": "Definite=Def|PronType=Prs", POS: DET},
|
||||
"DET__Definite=Ind|Gender=Fem|Number=Sing|PronType=Prs": {
|
||||
"morph": "Definite=Ind|Gender=Fem|Number=Sing|PronType=Prs",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Definite=Ind|Gender=Masc|Number=Sing|PronType=Prs": {
|
||||
"morph": "Definite=Ind|Gender=Masc|Number=Sing|PronType=Prs",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Definite=Ind|Gender=Neut|Number=Sing|PronType=Prs": {
|
||||
"morph": "Definite=Ind|Gender=Neut|Number=Sing|PronType=Prs",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Art": {
|
||||
"morph": "Gender=Fem|Number=Sing|PronType=Art",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Ind": {
|
||||
"morph": "Gender=Fem|Number=Sing|PronType=Ind",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Prs": {
|
||||
"morph": "Gender=Fem|Number=Sing|PronType=Prs",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Fem|Number=Sing|PronType=Tot": {
|
||||
"morph": "Gender=Fem|Number=Sing|PronType=Tot",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Masc|Number=Sing|Polarity=Neg|PronType=Neg": {
|
||||
"morph": "Gender=Masc|Number=Sing|Polarity=Neg|PronType=Neg",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Art": {
|
||||
"morph": "Gender=Masc|Number=Sing|PronType=Art",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Ind": {
|
||||
"morph": "Gender=Masc|Number=Sing|PronType=Ind",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Masc|Number=Sing|PronType=Tot": {
|
||||
"morph": "Gender=Masc|Number=Sing|PronType=Tot",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Neut|Number=Sing|Polarity=Neg|PronType=Neg": {
|
||||
"morph": "Gender=Neut|Number=Sing|Polarity=Neg|PronType=Neg",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Neut|Number=Sing|PronType=Art": {
|
||||
"morph": "Gender=Neut|Number=Sing|PronType=Art",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Neut|Number=Sing|PronType=Dem,Ind": {
|
||||
"morph": "Gender=Neut|Number=Sing|PronType=Dem,Ind",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Neut|Number=Sing|PronType=Ind": {
|
||||
"morph": "Gender=Neut|Number=Sing|PronType=Ind",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Gender=Neut|Number=Sing|PronType=Tot": {
|
||||
"morph": "Gender=Neut|Number=Sing|PronType=Tot",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Number=Plur|Polarity=Neg|PronType=Neg": {
|
||||
"morph": "Number=Plur|Polarity=Neg|PronType=Neg",
|
||||
POS: DET,
|
||||
},
|
||||
"DET__Number=Plur|PronType=Art": {"morph": "Number=Plur|PronType=Art", POS: DET},
|
||||
"DET__Number=Plur|PronType=Ind": {"morph": "Number=Plur|PronType=Ind", POS: DET},
|
||||
"DET__Number=Plur|PronType=Prs": {"morph": "Number=Plur|PronType=Prs", POS: DET},
|
||||
"DET__Number=Plur|PronType=Tot": {"morph": "Number=Plur|PronType=Tot", POS: DET},
|
||||
"DET__PronType=Ind": {"morph": "PronType=Ind", POS: DET},
|
||||
"DET__PronType=Prs": {"morph": "PronType=Prs", POS: DET},
|
||||
"NOUN__Abbr=Yes": {"morph": "Abbr=Yes", POS: NOUN},
|
||||
"NOUN__Abbr=Yes|Case=Gen": {"morph": "Abbr=Yes|Case=Gen", POS: NOUN},
|
||||
"NOUN__Abbr=Yes|Definite=Def,Ind|Gender=Masc|Number=Plur,Sing": {
|
||||
"morph": "Abbr=Yes|Definite=Def,Ind|Gender=Masc|Number=Plur,Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Abbr=Yes|Definite=Def,Ind|Gender=Masc|Number=Sing": {
|
||||
"morph": "Abbr=Yes|Definite=Def,Ind|Gender=Masc|Number=Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Abbr=Yes|Definite=Def,Ind|Gender=Neut|Number=Plur,Sing": {
|
||||
"morph": "Abbr=Yes|Definite=Def,Ind|Gender=Neut|Number=Plur,Sing",
|
||||
POS: NOUN,
|
||||
},
|
||||
"NOUN__Abbr=Yes|Gender=Masc": {"morph": "Abbr=Yes|Gender=Masc", POS: NOUN},
|
||||
"NUM__Case=Gen|Number=Plur|NumType=Card": {
|
||||
"morph": "Case=Gen|Number=Plur|NumType=Card",
|
||||
POS: NUM,
|
||||
},
|
||||
"NUM__Definite=Def|Number=Sing|NumType=Card": {
|
||||
"morph": "Definite=Def|Number=Sing|NumType=Card",
|
||||
POS: NUM,
|
||||
},
|
||||
"NUM__Definite=Def|NumType=Card": {"morph": "Definite=Def|NumType=Card", POS: NUM},
|
||||
"NUM__Gender=Fem|Number=Sing|NumType=Card": {
|
||||
"morph": "Gender=Fem|Number=Sing|NumType=Card",
|
||||
POS: NUM,
|
||||
},
|
||||
"NUM__Gender=Masc|Number=Sing|NumType=Card": {
|
||||
"morph": "Gender=Masc|Number=Sing|NumType=Card",
|
||||
POS: NUM,
|
||||
},
|
||||
"NUM__Gender=Neut|Number=Sing|NumType=Card": {
|
||||
"morph": "Gender=Neut|Number=Sing|NumType=Card",
|
||||
POS: NUM,
|
||||
},
|
||||
"NUM__Number=Plur|NumType=Card": {"morph": "Number=Plur|NumType=Card", POS: NUM},
|
||||
"NUM__Number=Sing|NumType=Card": {"morph": "Number=Sing|NumType=Card", POS: NUM},
|
||||
"NUM__NumType=Card": {"morph": "NumType=Card", POS: NUM},
|
||||
"PART__Polarity=Neg": {"morph": "Polarity=Neg", POS: PART},
|
||||
"PRON__Animacy=Hum|Case=Acc|Gender=Fem|Number=Sing|Person=3|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Acc|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Acc|Gender=Masc|Number=Sing|Person=3|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Acc|Gender=Masc|Number=Sing|Person=3|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Acc|Number=Plur|Person=1|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Acc|Number=Plur|Person=1|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Acc|Number=Plur|Person=2|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Acc|Number=Plur|Person=2|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Acc|Number=Sing|Person=1|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Acc|Number=Sing|Person=1|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Acc|Number=Sing|Person=2|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Acc|Number=Sing|Person=2|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Gen,Nom|Number=Sing|PronType=Art,Prs": {
|
||||
"morph": "Animacy=Hum|Case=Gen,Nom|Number=Sing|PronType=Art,Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Gen|Number=Sing|PronType=Art,Prs": {
|
||||
"morph": "Animacy=Hum|Case=Gen|Number=Sing|PronType=Art,Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Nom|Gender=Fem|Number=Sing|Person=3|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Nom|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Nom|Gender=Masc|Number=Sing|Person=3|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Nom|Gender=Masc|Number=Sing|Person=3|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Nom|Number=Plur|Person=1|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Nom|Number=Plur|Person=1|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Nom|Number=Plur|Person=2|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Nom|Number=Plur|Person=2|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Nom|Number=Sing|Person=1|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Nom|Number=Sing|Person=1|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Nom|Number=Sing|Person=2|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Nom|Number=Sing|Person=2|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Case=Nom|Number=Sing|PronType=Prs": {
|
||||
"morph": "Animacy=Hum|Case=Nom|Number=Sing|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Number=Plur|PronType=Rcp": {
|
||||
"morph": "Animacy=Hum|Number=Plur|PronType=Rcp",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Number=Sing|PronType=Art,Prs": {
|
||||
"morph": "Animacy=Hum|Number=Sing|PronType=Art,Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|Poss=Yes|PronType=Int": {
|
||||
"morph": "Animacy=Hum|Poss=Yes|PronType=Int",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Animacy=Hum|PronType=Int": {"morph": "Animacy=Hum|PronType=Int", POS: PRON},
|
||||
"PRON__Case=Acc|PronType=Prs|Reflex=Yes": {
|
||||
"morph": "Case=Acc|PronType=Prs|Reflex=Yes",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Gender=Fem,Masc|Number=Sing|Person=3|Polarity=Neg|PronType=Neg,Prs": {
|
||||
"morph": "Gender=Fem,Masc|Number=Sing|Person=3|Polarity=Neg|PronType=Neg,Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Gender=Fem,Masc|Number=Sing|Person=3|PronType=Ind,Prs": {
|
||||
"morph": "Gender=Fem,Masc|Number=Sing|Person=3|PronType=Ind,Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Gender=Fem,Masc|Number=Sing|Person=3|PronType=Prs,Tot": {
|
||||
"morph": "Gender=Fem,Masc|Number=Sing|Person=3|PronType=Prs,Tot",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Gender=Fem|Number=Sing|Poss=Yes|PronType=Prs": {
|
||||
"morph": "Gender=Fem|Number=Sing|Poss=Yes|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Gender=Masc|Number=Sing|Poss=Yes|PronType=Prs": {
|
||||
"morph": "Gender=Masc|Number=Sing|Poss=Yes|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Gender=Neut|Number=Sing|Person=3|PronType=Ind,Prs": {
|
||||
"morph": "Gender=Neut|Number=Sing|Person=3|PronType=Ind,Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Gender=Neut|Number=Sing|Poss=Yes|PronType=Prs": {
|
||||
"morph": "Gender=Neut|Number=Sing|Poss=Yes|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Number=Plur|Person=3|Polarity=Neg|PronType=Neg,Prs": {
|
||||
"morph": "Number=Plur|Person=3|Polarity=Neg|PronType=Neg,Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Number=Plur|Person=3|PronType=Ind,Prs": {
|
||||
"morph": "Number=Plur|Person=3|PronType=Ind,Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Number=Plur|Person=3|PronType=Prs,Tot": {
|
||||
"morph": "Number=Plur|Person=3|PronType=Prs,Tot",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Number=Plur|Poss=Yes|PronType=Prs": {
|
||||
"morph": "Number=Plur|Poss=Yes|PronType=Prs",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Number=Plur|Poss=Yes|PronType=Rcp": {
|
||||
"morph": "Number=Plur|Poss=Yes|PronType=Rcp",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__Number=Sing|Polarity=Neg|PronType=Neg": {
|
||||
"morph": "Number=Sing|Polarity=Neg|PronType=Neg",
|
||||
POS: PRON,
|
||||
},
|
||||
"PRON__PronType=Prs": {"morph": "PronType=Prs", POS: PRON},
|
||||
"PRON__PronType=Rel": {"morph": "PronType=Rel", POS: PRON},
|
||||
"PROPN__Abbr=Yes": {"morph": "Abbr=Yes", POS: PROPN},
|
||||
"PROPN__Abbr=Yes|Case=Gen": {"morph": "Abbr=Yes|Case=Gen", POS: PROPN},
|
||||
"VERB__Abbr=Yes|Mood=Ind|Tense=Pres|VerbForm=Fin": {
|
||||
"morph": "Abbr=Yes|Mood=Ind|Tense=Pres|VerbForm=Fin",
|
||||
POS: VERB,
|
||||
},
|
||||
"VERB__Definite=Ind|Number=Sing|VerbForm=Part": {
|
||||
"morph": "Definite=Ind|Number=Sing|VerbForm=Part",
|
||||
POS: VERB,
|
||||
},
|
||||
}
|
||||
|
|
@ -1,6 +1,5 @@
|
|||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .tag_map import TAG_MAP
|
||||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .punctuation import TOKENIZER_PREFIXES, TOKENIZER_INFIXES
|
||||
from .punctuation import TOKENIZER_SUFFIXES
|
||||
|
|
@ -22,7 +21,6 @@ class DutchDefaults(Language.Defaults):
|
|||
)
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
stop_words = STOP_WORDS
|
||||
tag_map = TAG_MAP
|
||||
prefixes = TOKENIZER_PREFIXES
|
||||
infixes = TOKENIZER_INFIXES
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
|
|
|
|||
File diff suppressed because it is too large
Load Diff
|
|
@ -1,6 +1,5 @@
|
|||
from .punctuation import TOKENIZER_PREFIXES, TOKENIZER_INFIXES
|
||||
from .punctuation import TOKENIZER_SUFFIXES
|
||||
from .tag_map import TAG_MAP
|
||||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .lemmatizer import PolishLemmatizer
|
||||
|
|
@ -25,7 +24,6 @@ class PolishDefaults(Language.Defaults):
|
|||
}
|
||||
tokenizer_exceptions = mod_base_exceptions
|
||||
stop_words = STOP_WORDS
|
||||
tag_map = TAG_MAP
|
||||
prefixes = TOKENIZER_PREFIXES
|
||||
infixes = TOKENIZER_INFIXES
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
|
|
|
|||
File diff suppressed because it is too large
Load Diff
|
|
@ -1,7 +1,6 @@
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .tag_map import TAG_MAP
|
||||
|
||||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
from .punctuation import TOKENIZER_INFIXES, TOKENIZER_PREFIXES
|
||||
|
|
@ -16,7 +15,6 @@ class PortugueseDefaults(Language.Defaults):
|
|||
lex_attr_getters.update(LEX_ATTRS)
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
stop_words = STOP_WORDS
|
||||
tag_map = TAG_MAP
|
||||
infixes = TOKENIZER_INFIXES
|
||||
prefixes = TOKENIZER_PREFIXES
|
||||
|
||||
|
|
|
|||
File diff suppressed because it is too large
Load Diff
|
|
@ -8,7 +8,6 @@ from ..norm_exceptions import BASE_NORMS
|
|||
from ...language import Language
|
||||
from ...attrs import LANG, NORM
|
||||
from ...util import update_exc, add_lookups
|
||||
from .tag_map import TAG_MAP
|
||||
|
||||
# Lemma data note:
|
||||
# Original pairs downloaded from http://www.lexiconista.com/datasets/lemmatization/
|
||||
|
|
@ -26,7 +25,6 @@ class RomanianDefaults(Language.Defaults):
|
|||
prefixes = TOKENIZER_PREFIXES
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
infixes = TOKENIZER_INFIXES
|
||||
tag_map = TAG_MAP
|
||||
|
||||
|
||||
class Romanian(Language):
|
||||
|
|
|
|||
File diff suppressed because it is too large
Load Diff
|
|
@ -1,7 +1,6 @@
|
|||
from .stop_words import STOP_WORDS
|
||||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .tag_map import TAG_MAP
|
||||
from .lemmatizer import RussianLemmatizer
|
||||
|
||||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
|
|
@ -17,7 +16,6 @@ class RussianDefaults(Language.Defaults):
|
|||
lex_attr_getters[LANG] = lambda text: "ru"
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
stop_words = STOP_WORDS
|
||||
tag_map = TAG_MAP
|
||||
|
||||
@classmethod
|
||||
def create_lemmatizer(cls, nlp=None, lookups=None):
|
||||
|
|
|
|||
|
|
@ -1,743 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, SYM, ADJ, NUM, DET, ADV, ADP, X, VERB, NOUN
|
||||
from ...symbols import PROPN, PART, INTJ, PRON, SCONJ, AUX, CCONJ
|
||||
|
||||
# fmt: off
|
||||
TAG_MAP = {
|
||||
'ADJ__Animacy=Anim|Case=Acc|Degree=Pos|Gender=Masc|Number=Sing': {POS: ADJ, 'Animacy': 'Anim', 'Case': 'Acc', 'Degree': 'Pos', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Animacy=Anim|Case=Acc|Degree=Pos|Number=Plur': {POS: ADJ, 'Animacy': 'Anim', 'Case': 'Acc', 'Degree': 'Pos', 'Number': 'Plur'},
|
||||
'ADJ__Animacy=Anim|Case=Acc|Degree=Sup|Gender=Masc|Number=Sing': {POS: ADJ, 'Animacy': 'Anim', 'Case': 'Acc', 'Degree': 'Sup', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Animacy=Anim|Case=Nom|Degree=Pos|Number=Plur': {POS: ADJ, 'Animacy': 'Anim', 'Case': 'Nom', 'Degree': 'Pos', 'Number': 'Plur'},
|
||||
'ADJ__Animacy=Inan|Case=Acc|Degree=Pos|Gender=Masc|Number=Sing': {POS: ADJ, 'Animacy': 'Inan', 'Case': 'Acc', 'Degree': 'Pos', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Animacy=Inan|Case=Acc|Degree=Pos|Gender=Neut|Number=Sing': {POS: ADJ, 'Animacy': 'Inan', 'Case': 'Acc', 'Degree': 'Pos', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Animacy=Inan|Case=Acc|Degree=Pos|Number=Plur': {POS: ADJ, 'Animacy': 'Inan', 'Case': 'Acc', 'Degree': 'Pos', 'Number': 'Plur'},
|
||||
'ADJ__Animacy=Inan|Case=Acc|Degree=Sup|Gender=Masc|Number=Sing': {POS: ADJ, 'Animacy': 'Inan', 'Case': 'Acc', 'Degree': 'Sup', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Animacy=Inan|Case=Acc|Degree=Sup|Number=Plur': {POS: ADJ, 'Animacy': 'Inan', 'Case': 'Acc', 'Degree': 'Sup', 'Number': 'Plur'},
|
||||
'ADJ__Animacy=Inan|Case=Acc|Gender=Fem|Number=Sing': {POS: ADJ, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Animacy=Inan|Case=Nom|Degree=Pos|Gender=Fem|Number=Sing': {POS: ADJ, 'Animacy': 'Inan', 'Case': 'Nom', 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Case=Acc|Degree=Pos|Gender=Fem|Number=Sing': {POS: ADJ, 'Case': 'Acc', 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Case=Acc|Degree=Pos|Gender=Neut|Number=Sing': {POS: ADJ, 'Case': 'Acc', 'Degree': 'Pos', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Case=Acc|Degree=Sup|Gender=Fem|Number=Sing': {POS: ADJ, 'Case': 'Acc', 'Degree': 'Sup', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Case=Acc|Degree=Sup|Gender=Neut|Number=Sing': {POS: ADJ, 'Case': 'Acc', 'Degree': 'Sup', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Case=Dat|Degree=Pos|Gender=Fem|Number=Sing': {POS: ADJ, 'Case': 'Dat', 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Case=Dat|Degree=Pos|Gender=Masc|Number=Sing': {POS: ADJ, 'Case': 'Dat', 'Degree': 'Pos', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Case=Dat|Degree=Pos|Gender=Neut|Number=Sing': {POS: ADJ, 'Case': 'Dat', 'Degree': 'Pos', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Case=Dat|Degree=Pos|Number=Plur': {POS: ADJ, 'Case': 'Dat', 'Degree': 'Pos', 'Number': 'Plur'},
|
||||
'ADJ__Case=Dat|Degree=Sup|Gender=Masc|Number=Sing': {POS: ADJ, 'Case': 'Dat', 'Degree': 'Sup', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Case=Dat|Degree=Sup|Gender=Neut|Number=Sing': {POS: ADJ, 'Case': 'Dat', 'Degree': 'Sup', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Case=Dat|Degree=Sup|Number=Plur': {POS: ADJ, 'Case': 'Dat', 'Degree': 'Sup', 'Number': 'Plur'},
|
||||
'ADJ__Case=Gen|Degree=Pos|Gender=Fem|Number=Sing': {POS: ADJ, 'Case': 'Gen', 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Case=Gen|Degree=Pos|Gender=Fem|Number=Sing|Variant=Short': {POS: ADJ, 'Case': 'Gen', 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing', 'Variant': 'Short'},
|
||||
'ADJ__Case=Gen|Degree=Pos|Gender=Masc|Number=Sing': {POS: ADJ, 'Case': 'Gen', 'Degree': 'Pos', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Case=Gen|Degree=Pos|Gender=Neut|Number=Sing': {POS: ADJ, 'Case': 'Gen', 'Degree': 'Pos', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Case=Gen|Degree=Pos|Number=Plur': {POS: ADJ, 'Case': 'Gen', 'Degree': 'Pos', 'Number': 'Plur'},
|
||||
'ADJ__Case=Gen|Degree=Sup|Gender=Fem|Number=Sing': {POS: ADJ, 'Case': 'Gen', 'Degree': 'Sup', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Case=Gen|Degree=Sup|Gender=Masc|Number=Sing': {POS: ADJ, 'Case': 'Gen', 'Degree': 'Sup', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Case=Gen|Degree=Sup|Gender=Neut|Number=Sing': {POS: ADJ, 'Case': 'Gen', 'Degree': 'Sup', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Case=Gen|Degree=Sup|Number=Plur': {POS: ADJ, 'Case': 'Gen', 'Degree': 'Sup', 'Number': 'Plur'},
|
||||
'ADJ__Case=Ins|Degree=Pos|Gender=Fem|Number=Sing': {POS: ADJ, 'Case': 'Ins', 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Case=Ins|Degree=Pos|Gender=Masc|Number=Sing': {POS: ADJ, 'Case': 'Ins', 'Degree': 'Pos', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Case=Ins|Degree=Pos|Gender=Neut|Number=Sing': {POS: ADJ, 'Case': 'Ins', 'Degree': 'Pos', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Case=Ins|Degree=Pos|Number=Plur': {POS: ADJ, 'Case': 'Ins', 'Degree': 'Pos', 'Number': 'Plur'},
|
||||
'ADJ__Case=Ins|Degree=Sup|Gender=Fem|Number=Sing': {POS: ADJ, 'Case': 'Ins', 'Degree': 'Sup', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Case=Ins|Degree=Sup|Gender=Masc|Number=Sing': {POS: ADJ, 'Case': 'Ins', 'Degree': 'Sup', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Case=Ins|Degree=Sup|Gender=Neut|Number=Sing': {POS: ADJ, 'Case': 'Ins', 'Degree': 'Sup', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Case=Ins|Degree=Sup|Number=Plur': {POS: ADJ, 'Case': 'Ins', 'Degree': 'Sup', 'Number': 'Plur'},
|
||||
'ADJ__Case=Loc|Degree=Pos|Gender=Fem|Number=Sing': {POS: ADJ, 'Case': 'Loc', 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Case=Loc|Degree=Pos|Gender=Masc|Number=Sing': {POS: ADJ, 'Case': 'Loc', 'Degree': 'Pos', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Case=Loc|Degree=Pos|Gender=Neut|Number=Sing': {POS: ADJ, 'Case': 'Loc', 'Degree': 'Pos', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Case=Loc|Degree=Pos|Number=Plur': {POS: ADJ, 'Case': 'Loc', 'Degree': 'Pos', 'Number': 'Plur'},
|
||||
'ADJ__Case=Loc|Degree=Sup|Gender=Fem|Number=Sing': {POS: ADJ, 'Case': 'Loc', 'Degree': 'Sup', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Case=Loc|Degree=Sup|Gender=Masc|Number=Sing': {POS: ADJ, 'Case': 'Loc', 'Degree': 'Sup', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Case=Loc|Degree=Sup|Gender=Neut|Number=Sing': {POS: ADJ, 'Case': 'Loc', 'Degree': 'Sup', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Case=Loc|Degree=Sup|Number=Plur': {POS: ADJ, 'Case': 'Loc', 'Degree': 'Sup', 'Number': 'Plur'},
|
||||
'ADJ__Case=Nom|Degree=Pos|Gender=Fem|Number=Sing': {POS: ADJ, 'Case': 'Nom', 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Case=Nom|Degree=Pos|Gender=Masc|Number=Sing': {POS: ADJ, 'Case': 'Nom', 'Degree': 'Pos', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Case=Nom|Degree=Pos|Gender=Neut|Number=Sing': {POS: ADJ, 'Case': 'Nom', 'Degree': 'Pos', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Case=Nom|Degree=Pos|Number=Plur': {POS: ADJ, 'Case': 'Nom', 'Degree': 'Pos', 'Number': 'Plur'},
|
||||
'ADJ__Case=Nom|Degree=Sup|Gender=Fem|Number=Sing': {POS: ADJ, 'Case': 'Nom', 'Degree': 'Sup', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'ADJ__Case=Nom|Degree=Sup|Gender=Masc|Number=Sing': {POS: ADJ, 'Case': 'Nom', 'Degree': 'Sup', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'ADJ__Case=Nom|Degree=Sup|Gender=Neut|Number=Sing': {POS: ADJ, 'Case': 'Nom', 'Degree': 'Sup', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'ADJ__Case=Nom|Degree=Sup|Number=Plur': {POS: ADJ, 'Case': 'Nom', 'Degree': 'Sup', 'Number': 'Plur'},
|
||||
'ADJ__Degree=Cmp': {POS: ADJ, 'Degree': 'Cmp'},
|
||||
'ADJ__Degree=Pos': {POS: ADJ, 'Degree': 'Pos'},
|
||||
'ADJ__Degree=Pos|Gender=Fem|Number=Sing|Variant=Short': {POS: ADJ, 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing', 'Variant': 'Short'},
|
||||
'ADJ__Degree=Pos|Gender=Masc|Number=Sing|Variant=Short': {POS: ADJ, 'Degree': 'Pos', 'Gender': 'Masc', 'Number': 'Sing', 'Variant': 'Short'},
|
||||
'ADJ__Degree=Pos|Gender=Neut|Number=Sing|Variant=Short': {POS: ADJ, 'Degree': 'Pos', 'Gender': 'Neut', 'Number': 'Sing', 'Variant': 'Short'},
|
||||
'ADJ__Degree=Pos|Number=Plur|Variant=Short': {POS: ADJ, 'Degree': 'Pos', 'Number': 'Plur', 'Variant': 'Short'},
|
||||
'ADJ__Foreign=Yes': {POS: ADJ, 'Foreign': 'Yes'},
|
||||
'ADJ___': {POS: ADJ},
|
||||
'ADJ': {POS: ADJ},
|
||||
'ADP___': {POS: ADP},
|
||||
'ADP': {POS: ADP},
|
||||
'ADV__Degree=Cmp': {POS: ADV, 'Degree': 'Cmp'},
|
||||
'ADV__Degree=Pos': {POS: ADV, 'Degree': 'Pos'},
|
||||
'ADV__Polarity=Neg': {POS: ADV, 'Polarity': 'Neg'},
|
||||
'AUX__Aspect=Imp|Case=Loc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Case=Nom|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Case=Nom|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Case': 'Nom', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Gender=Fem|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Gender': 'Fem', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Gender=Masc|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Gender': 'Masc', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Gender=Neut|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Gender': 'Neut', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Mood=Imp|Number=Plur|Person=2|VerbForm=Fin|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Mood': 'Imp', 'Number': 'Plur', 'Person': '2', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Mood=Imp|Number=Sing|Person=2|VerbForm=Fin|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Mood': 'Imp', 'Number': 'Sing', 'Person': '2', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '1', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '2', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '3', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Mood=Ind|Number=Plur|Tense=Past|VerbForm=Fin|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '1', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Mood=Ind|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '2', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '3', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|Tense=Pres|VerbForm=Conv|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'Tense': 'Pres', 'VerbForm': 'Conv', 'Voice': 'Act'},
|
||||
'AUX__Aspect=Imp|VerbForm=Inf|Voice=Act': {POS: AUX, 'Aspect': 'Imp', 'VerbForm': 'Inf', 'Voice': 'Act'},
|
||||
'CCONJ___': {POS: CCONJ},
|
||||
'CCONJ': {POS: CCONJ},
|
||||
'DET__Animacy=Inan|Case=Acc|Gender=Masc|Number=Sing': {POS: DET, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'DET__Animacy=Inan|Case=Acc|Gender=Neut|Number=Sing': {POS: DET, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'DET__Animacy=Inan|Case=Gen|Gender=Fem|Number=Sing': {POS: DET, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'DET__Animacy=Inan|Case=Gen|Number=Plur': {POS: DET, 'Animacy': 'Inan', 'Case': 'Gen', 'Number': 'Plur'},
|
||||
'DET__Case=Acc|Degree=Pos|Number=Plur': {POS: DET, 'Case': 'Acc', 'Degree': 'Pos', 'Number': 'Plur'},
|
||||
'DET__Case=Acc|Gender=Fem|Number=Sing': {POS: DET, 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'DET__Case=Acc|Gender=Masc|Number=Sing': {POS: DET, 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'DET__Case=Acc|Gender=Neut|Number=Sing': {POS: DET, 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'DET__Case=Acc|Number=Plur': {POS: DET, 'Case': 'Acc', 'Number': 'Plur'},
|
||||
'DET__Case=Dat|Gender=Fem|Number=Sing': {POS: DET, 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'DET__Case=Dat|Gender=Masc|Number=Plur': {POS: DET, 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'DET__Case=Dat|Gender=Masc|Number=Sing': {POS: DET, 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'DET__Case=Dat|Gender=Neut|Number=Sing': {POS: DET, 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'DET__Case=Dat|Number=Plur': {POS: DET, 'Case': 'Dat', 'Number': 'Plur'},
|
||||
'DET__Case=Gen|Gender=Fem|Number=Sing': {POS: DET, 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'DET__Case=Gen|Gender=Masc|Number=Sing': {POS: DET, 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'DET__Case=Gen|Gender=Neut|Number=Sing': {POS: DET, 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'DET__Case=Gen|Number=Plur': {POS: DET, 'Case': 'Gen', 'Number': 'Plur'},
|
||||
'DET__Case=Ins|Gender=Fem|Number=Sing': {POS: DET, 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'DET__Case=Ins|Gender=Masc|Number=Sing': {POS: DET, 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'DET__Case=Ins|Gender=Neut|Number=Sing': {POS: DET, 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'DET__Case=Ins|Number=Plur': {POS: DET, 'Case': 'Ins', 'Number': 'Plur'},
|
||||
'DET__Case=Loc|Gender=Fem|Number=Sing': {POS: DET, 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'DET__Case=Loc|Gender=Masc|Number=Sing': {POS: DET, 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'DET__Case=Loc|Gender=Neut|Number=Sing': {POS: DET, 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'DET__Case=Loc|Number=Plur': {POS: DET, 'Case': 'Loc', 'Number': 'Plur'},
|
||||
'DET__Case=Nom|Gender=Fem|Number=Sing': {POS: DET, 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'DET__Case=Nom|Gender=Masc|Number=Plur': {POS: DET, 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'DET__Case=Nom|Gender=Masc|Number=Sing': {POS: DET, 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'DET__Case=Nom|Gender=Neut|Number=Sing': {POS: DET, 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'DET__Case=Nom|Number=Plur': {POS: DET, 'Case': 'Nom', 'Number': 'Plur'},
|
||||
'DET__Gender=Masc|Number=Sing': {POS: DET, 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'INTJ___': {POS: INTJ},
|
||||
'INTJ': {POS: INTJ},
|
||||
'NOUN__Animacy=Anim|Case=Acc|Gender=Fem|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Acc|Gender=Fem|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Acc|Gender=Masc|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Acc|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Acc|Gender=Neut|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Acc|Gender=Neut|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Acc|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Acc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Dat|Gender=Fem|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Dat|Gender=Fem|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Dat|Gender=Masc|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Dat|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Dat|Gender=Neut|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Dat|Gender=Neut|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Dat|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Dat', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Gen|Gender=Fem|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Gen|Gender=Fem|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Gen|Gender=Masc|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Gen|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Gen|Gender=Neut|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Gen|Gender=Neut|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Gen|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Gen', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Ins|Gender=Fem|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Ins|Gender=Fem|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Ins|Gender=Masc|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Ins|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Ins|Gender=Neut|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Ins|Gender=Neut|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Ins|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Ins', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Loc|Gender=Fem|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Loc|Gender=Fem|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Loc|Gender=Masc|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Loc|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Loc|Gender=Neut|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Loc|Gender=Neut|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Loc|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Loc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Nom|Gender=Fem|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Nom|Gender=Fem|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Nom|Gender=Masc|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Nom|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Nom|Gender=Neut|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Nom|Gender=Neut|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Anim|Case=Nom|Number=Plur': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Nom', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Anim|Case=Voc|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Anim', 'Case': 'Voc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Acc|Gender=Fem|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Acc|Gender=Fem|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Acc|Gender=Masc|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Acc|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Acc|Gender=Neut|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Acc|Gender=Neut|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Acc|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Acc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Dat|Gender=Fem|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Dat|Gender=Fem|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Dat|Gender=Masc|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Dat|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Dat|Gender=Neut|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Dat|Gender=Neut|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Dat|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Dat', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Gen|Gender=Fem|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Gen|Gender=Fem|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Gen|Gender=Masc|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Gen|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Gen|Gender=Neut|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Gen|Gender=Neut|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Gen|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Gen', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Ins|Gender=Fem|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Ins|Gender=Fem|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Ins|Gender=Masc|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Ins|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Ins|Gender=Neut|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Ins|Gender=Neut|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Ins|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Ins', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Loc|Gender=Fem|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Loc|Gender=Fem|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Loc|Gender=Masc|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Loc|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Loc|Gender=Neut|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Loc|Gender=Neut|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Loc|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Loc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Nom|Gender=Fem|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Nom|Gender=Fem|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Nom|Gender=Masc|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Nom|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Nom|Gender=Neut|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Nom|Gender=Neut|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Case=Nom|Number=Plur': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Nom', 'Number': 'Plur'},
|
||||
'NOUN__Animacy=Inan|Case=Par|Gender=Masc|Number=Sing': {POS: NOUN, 'Animacy': 'Inan', 'Case': 'Par', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'NOUN__Animacy=Inan|Gender=Fem': {POS: NOUN, 'Animacy': 'Inan', 'Gender': 'Fem'},
|
||||
'NOUN__Animacy=Inan|Gender=Masc': {POS: NOUN, 'Animacy': 'Inan', 'Gender': 'Masc'},
|
||||
'NOUN__Animacy=Inan|Gender=Neut': {POS: NOUN, 'Animacy': 'Inan', 'Gender': 'Neut'},
|
||||
'NOUN__Case=Gen|Degree=Pos|Gender=Fem|Number=Sing': {POS: NOUN, 'Case': 'Gen', 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'NOUN__Foreign=Yes': {POS: NOUN, 'Foreign': 'Yes'},
|
||||
'NOUN___': {POS: NOUN},
|
||||
'NOUN': {POS: NOUN},
|
||||
'NUM__Animacy=Anim|Case=Acc': {POS: NUM, 'Animacy': 'Anim', 'Case': 'Acc'},
|
||||
'NUM__Animacy=Anim|Case=Acc|Gender=Fem': {POS: NUM, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Fem'},
|
||||
'NUM__Animacy=Anim|Case=Acc|Gender=Masc': {POS: NUM, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Masc'},
|
||||
'NUM__Animacy=Inan|Case=Acc': {POS: NUM, 'Animacy': 'Inan', 'Case': 'Acc'},
|
||||
'NUM__Animacy=Inan|Case=Acc|Gender=Fem': {POS: NUM, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Fem'},
|
||||
'NUM__Animacy=Inan|Case=Acc|Gender=Masc': {POS: NUM, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Masc'},
|
||||
'NUM__Case=Acc': {POS: NUM, 'Case': 'Acc'},
|
||||
'NUM__Case=Acc|Gender=Fem': {POS: NUM, 'Case': 'Acc', 'Gender': 'Fem'},
|
||||
'NUM__Case=Acc|Gender=Masc': {POS: NUM, 'Case': 'Acc', 'Gender': 'Masc'},
|
||||
'NUM__Case=Acc|Gender=Neut': {POS: NUM, 'Case': 'Acc', 'Gender': 'Neut'},
|
||||
'NUM__Case=Dat': {POS: NUM, 'Case': 'Dat'},
|
||||
'NUM__Case=Dat|Gender=Fem': {POS: NUM, 'Case': 'Dat', 'Gender': 'Fem'},
|
||||
'NUM__Case=Dat|Gender=Masc': {POS: NUM, 'Case': 'Dat', 'Gender': 'Masc'},
|
||||
'NUM__Case=Dat|Gender=Neut': {POS: NUM, 'Case': 'Dat', 'Gender': 'Neut'},
|
||||
'NUM__Case=Gen': {POS: NUM, 'Case': 'Gen'},
|
||||
'NUM__Case=Gen|Gender=Fem': {POS: NUM, 'Case': 'Gen', 'Gender': 'Fem'},
|
||||
'NUM__Case=Gen|Gender=Masc': {POS: NUM, 'Case': 'Gen', 'Gender': 'Masc'},
|
||||
'NUM__Case=Gen|Gender=Neut': {POS: NUM, 'Case': 'Gen', 'Gender': 'Neut'},
|
||||
'NUM__Case=Ins': {POS: NUM, 'Case': 'Ins'},
|
||||
'NUM__Case=Ins|Gender=Fem': {POS: NUM, 'Case': 'Ins', 'Gender': 'Fem'},
|
||||
'NUM__Case=Ins|Gender=Masc': {POS: NUM, 'Case': 'Ins', 'Gender': 'Masc'},
|
||||
'NUM__Case=Ins|Gender=Neut': {POS: NUM, 'Case': 'Ins', 'Gender': 'Neut'},
|
||||
'NUM__Case=Loc': {POS: NUM, 'Case': 'Loc'},
|
||||
'NUM__Case=Loc|Gender=Fem': {POS: NUM, 'Case': 'Loc', 'Gender': 'Fem'},
|
||||
'NUM__Case=Loc|Gender=Masc': {POS: NUM, 'Case': 'Loc', 'Gender': 'Masc'},
|
||||
'NUM__Case=Loc|Gender=Neut': {POS: NUM, 'Case': 'Loc', 'Gender': 'Neut'},
|
||||
'NUM__Case=Nom': {POS: NUM, 'Case': 'Nom'},
|
||||
'NUM__Case=Nom|Gender=Fem': {POS: NUM, 'Case': 'Nom', 'Gender': 'Fem'},
|
||||
'NUM__Case=Nom|Gender=Masc': {POS: NUM, 'Case': 'Nom', 'Gender': 'Masc'},
|
||||
'NUM__Case=Nom|Gender=Neut': {POS: NUM, 'Case': 'Nom', 'Gender': 'Neut'},
|
||||
'NUM___': {POS: NUM},
|
||||
'NUM': {POS: NUM},
|
||||
'PART__Mood=Cnd': {POS: PART, 'Mood': 'Cnd'},
|
||||
'PART__Polarity=Neg': {POS: PART, 'Polarity': 'Neg'},
|
||||
'PART___': {POS: PART},
|
||||
'PART': {POS: PART},
|
||||
'PRON__Animacy=Anim|Case=Acc|Gender=Masc|Number=Plur': {POS: PRON, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PRON__Animacy=Anim|Case=Acc|Number=Plur': {POS: PRON, 'Animacy': 'Anim', 'Case': 'Acc', 'Number': 'Plur'},
|
||||
'PRON__Animacy=Anim|Case=Dat|Gender=Masc|Number=Sing': {POS: PRON, 'Animacy': 'Anim', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PRON__Animacy=Anim|Case=Dat|Number=Plur': {POS: PRON, 'Animacy': 'Anim', 'Case': 'Dat', 'Number': 'Plur'},
|
||||
'PRON__Animacy=Anim|Case=Gen|Number=Plur': {POS: PRON, 'Animacy': 'Anim', 'Case': 'Gen', 'Number': 'Plur'},
|
||||
'PRON__Animacy=Anim|Case=Ins|Gender=Masc|Number=Sing': {POS: PRON, 'Animacy': 'Anim', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PRON__Animacy=Anim|Case=Ins|Number=Plur': {POS: PRON, 'Animacy': 'Anim', 'Case': 'Ins', 'Number': 'Plur'},
|
||||
'PRON__Animacy=Anim|Case=Loc|Number=Plur': {POS: PRON, 'Animacy': 'Anim', 'Case': 'Loc', 'Number': 'Plur'},
|
||||
'PRON__Animacy=Anim|Case=Nom|Gender=Masc|Number=Plur': {POS: PRON, 'Animacy': 'Anim', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PRON__Animacy=Anim|Case=Nom|Number=Plur': {POS: PRON, 'Animacy': 'Anim', 'Case': 'Nom', 'Number': 'Plur'},
|
||||
'PRON__Animacy=Anim|Gender=Masc|Number=Plur': {POS: PRON, 'Animacy': 'Anim', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PRON__Animacy=Inan|Case=Acc|Gender=Masc|Number=Sing': {POS: PRON, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PRON__Animacy=Inan|Case=Acc|Gender=Neut|Number=Sing': {POS: PRON, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PRON__Animacy=Inan|Case=Dat|Gender=Neut|Number=Sing': {POS: PRON, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PRON__Animacy=Inan|Case=Gen|Gender=Masc|Number=Sing': {POS: PRON, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PRON__Animacy=Inan|Case=Gen|Gender=Neut|Number=Sing': {POS: PRON, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PRON__Animacy=Inan|Case=Ins|Gender=Fem|Number=Sing': {POS: PRON, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PRON__Animacy=Inan|Case=Ins|Gender=Neut|Number=Sing': {POS: PRON, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PRON__Animacy=Inan|Case=Loc|Gender=Neut|Number=Sing': {POS: PRON, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PRON__Animacy=Inan|Case=Nom|Gender=Neut|Number=Sing': {POS: PRON, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PRON__Animacy=Inan|Gender=Neut|Number=Sing': {POS: PRON, 'Animacy': 'Inan', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PRON__Case=Acc': {POS: PRON, 'Case': 'Acc'},
|
||||
'PRON__Case=Acc|Gender=Fem|Number=Sing|Person=3': {POS: PRON, 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Acc|Gender=Masc|Number=Sing|Person=3': {POS: PRON, 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Acc|Gender=Neut|Number=Sing|Person=3': {POS: PRON, 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Acc|Number=Plur|Person=1': {POS: PRON, 'Case': 'Acc', 'Number': 'Plur', 'Person': '1'},
|
||||
'PRON__Case=Acc|Number=Plur|Person=2': {POS: PRON, 'Case': 'Acc', 'Number': 'Plur', 'Person': '2'},
|
||||
'PRON__Case=Acc|Number=Plur|Person=3': {POS: PRON, 'Case': 'Acc', 'Number': 'Plur', 'Person': '3'},
|
||||
'PRON__Case=Acc|Number=Sing|Person=1': {POS: PRON, 'Case': 'Acc', 'Number': 'Sing', 'Person': '1'},
|
||||
'PRON__Case=Acc|Number=Sing|Person=2': {POS: PRON, 'Case': 'Acc', 'Number': 'Sing', 'Person': '2'},
|
||||
'PRON__Case=Dat': {POS: PRON, 'Case': 'Dat'},
|
||||
'PRON__Case=Dat|Gender=Fem|Number=Sing|Person=3': {POS: PRON, 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Dat|Gender=Masc|Number=Sing|Person=3': {POS: PRON, 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Dat|Gender=Neut|Number=Sing|Person=3': {POS: PRON, 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Dat|Number=Plur|Person=1': {POS: PRON, 'Case': 'Dat', 'Number': 'Plur', 'Person': '1'},
|
||||
'PRON__Case=Dat|Number=Plur|Person=2': {POS: PRON, 'Case': 'Dat', 'Number': 'Plur', 'Person': '2'},
|
||||
'PRON__Case=Dat|Number=Plur|Person=3': {POS: PRON, 'Case': 'Dat', 'Number': 'Plur', 'Person': '3'},
|
||||
'PRON__Case=Dat|Number=Sing|Person=1': {POS: PRON, 'Case': 'Dat', 'Number': 'Sing', 'Person': '1'},
|
||||
'PRON__Case=Dat|Number=Sing|Person=2': {POS: PRON, 'Case': 'Dat', 'Number': 'Sing', 'Person': '2'},
|
||||
'PRON__Case=Gen': {POS: PRON, 'Case': 'Gen'},
|
||||
'PRON__Case=Gen|Gender=Fem|Number=Sing|Person=3': {POS: PRON, 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Gen|Gender=Masc|Number=Sing|Person=3': {POS: PRON, 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Gen|Gender=Neut|Number=Sing|Person=3': {POS: PRON, 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Gen|Number=Plur|Person=1': {POS: PRON, 'Case': 'Gen', 'Number': 'Plur', 'Person': '1'},
|
||||
'PRON__Case=Gen|Number=Plur|Person=2': {POS: PRON, 'Case': 'Gen', 'Number': 'Plur', 'Person': '2'},
|
||||
'PRON__Case=Gen|Number=Plur|Person=3': {POS: PRON, 'Case': 'Gen', 'Number': 'Plur', 'Person': '3'},
|
||||
'PRON__Case=Gen|Number=Sing|Person=1': {POS: PRON, 'Case': 'Gen', 'Number': 'Sing', 'Person': '1'},
|
||||
'PRON__Case=Gen|Number=Sing|Person=2': {POS: PRON, 'Case': 'Gen', 'Number': 'Sing', 'Person': '2'},
|
||||
'PRON__Case=Ins': {POS: PRON, 'Case': 'Ins'},
|
||||
'PRON__Case=Ins|Gender=Fem|Number=Sing|Person=3': {POS: PRON, 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Ins|Gender=Masc|Number=Sing|Person=3': {POS: PRON, 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Ins|Gender=Neut|Number=Sing|Person=3': {POS: PRON, 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Ins|Number=Plur|Person=1': {POS: PRON, 'Case': 'Ins', 'Number': 'Plur', 'Person': '1'},
|
||||
'PRON__Case=Ins|Number=Plur|Person=2': {POS: PRON, 'Case': 'Ins', 'Number': 'Plur', 'Person': '2'},
|
||||
'PRON__Case=Ins|Number=Plur|Person=3': {POS: PRON, 'Case': 'Ins', 'Number': 'Plur', 'Person': '3'},
|
||||
'PRON__Case=Ins|Number=Sing|Person=1': {POS: PRON, 'Case': 'Ins', 'Number': 'Sing', 'Person': '1'},
|
||||
'PRON__Case=Ins|Number=Sing|Person=2': {POS: PRON, 'Case': 'Ins', 'Number': 'Sing', 'Person': '2'},
|
||||
'PRON__Case=Loc': {POS: PRON, 'Case': 'Loc'},
|
||||
'PRON__Case=Loc|Gender=Fem|Number=Sing|Person=3': {POS: PRON, 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Loc|Gender=Masc|Number=Sing|Person=3': {POS: PRON, 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Loc|Gender=Neut|Number=Sing|Person=3': {POS: PRON, 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Loc|Number=Plur|Person=1': {POS: PRON, 'Case': 'Loc', 'Number': 'Plur', 'Person': '1'},
|
||||
'PRON__Case=Loc|Number=Plur|Person=2': {POS: PRON, 'Case': 'Loc', 'Number': 'Plur', 'Person': '2'},
|
||||
'PRON__Case=Loc|Number=Plur|Person=3': {POS: PRON, 'Case': 'Loc', 'Number': 'Plur', 'Person': '3'},
|
||||
'PRON__Case=Loc|Number=Sing|Person=1': {POS: PRON, 'Case': 'Loc', 'Number': 'Sing', 'Person': '1'},
|
||||
'PRON__Case=Loc|Number=Sing|Person=2': {POS: PRON, 'Case': 'Loc', 'Number': 'Sing', 'Person': '2'},
|
||||
'PRON__Case=Nom': {POS: PRON, 'Case': 'Nom'},
|
||||
'PRON__Case=Nom|Gender=Fem|Number=Sing|Person=3': {POS: PRON, 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Nom|Gender=Masc|Number=Sing|Person=3': {POS: PRON, 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Nom|Gender=Neut|Number=Sing|Person=3': {POS: PRON, 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing', 'Person': '3'},
|
||||
'PRON__Case=Nom|Number=Plur|Person=1': {POS: PRON, 'Case': 'Nom', 'Number': 'Plur', 'Person': '1'},
|
||||
'PRON__Case=Nom|Number=Plur|Person=2': {POS: PRON, 'Case': 'Nom', 'Number': 'Plur', 'Person': '2'},
|
||||
'PRON__Case=Nom|Number=Plur|Person=3': {POS: PRON, 'Case': 'Nom', 'Number': 'Plur', 'Person': '3'},
|
||||
'PRON__Case=Nom|Number=Sing|Person=1': {POS: PRON, 'Case': 'Nom', 'Number': 'Sing', 'Person': '1'},
|
||||
'PRON__Case=Nom|Number=Sing|Person=2': {POS: PRON, 'Case': 'Nom', 'Number': 'Sing', 'Person': '2'},
|
||||
'PRON__Number=Sing|Person=1': {POS: PRON, 'Number': 'Sing', 'Person': '1'},
|
||||
'PRON___': {POS: PRON},
|
||||
'PRON': {POS: PRON},
|
||||
'PROPN__Animacy=Anim|Case=Acc|Gender=Fem|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Acc|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Acc|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Acc|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Acc|Gender=Neut|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Dat|Gender=Fem|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Dat|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Dat|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Dat|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Dat|Gender=Neut|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Gen|Foreign=Yes|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Gen', 'Foreign': 'Yes', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Gen|Gender=Fem|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Gen|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Gen|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Gen|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Ins|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Ins|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Ins|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Ins|Gender=Neut|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Loc|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Loc|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Loc|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Nom|Foreign=Yes|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Nom', 'Foreign': 'Yes', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Nom|Gender=Fem|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Nom|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Nom|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Nom|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Case=Nom|Gender=Neut|Number=Plur': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Anim|Case=Voc|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Case': 'Voc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Anim|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Anim', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Acc|Gender=Fem|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Acc|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Acc|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Acc|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Acc|Gender=Neut|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Acc|Gender=Neut|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Acc|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Acc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Dat|Gender=Fem|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Dat|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Dat|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Dat|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Dat|Gender=Neut|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Dat|Gender=Neut|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Dat|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Dat', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Gen|Foreign=Yes|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Gen', 'Foreign': 'Yes', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Gen|Gender=Fem|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Gen|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Gen|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Gen|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Gen|Gender=Neut|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Gen|Gender=Neut|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Gen|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Gen', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Ins|Gender=Fem|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Ins|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Ins|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Ins|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Ins|Gender=Neut|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Ins|Gender=Neut|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Ins|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Ins', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Loc|Gender=Fem|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Loc|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Loc|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Loc|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Loc|Gender=Neut|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Loc|Gender=Neut|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Loc|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Loc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Nom|Foreign=Yes|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Nom', 'Foreign': 'Yes', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Nom|Foreign=Yes|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Nom', 'Foreign': 'Yes', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Nom|Foreign=Yes|Gender=Neut|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Nom', 'Foreign': 'Yes', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Nom|Gender=Fem|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Nom|Gender=Fem|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Nom|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Nom|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Nom|Gender=Neut|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Nom|Gender=Neut|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Case=Nom|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Nom', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Case=Par|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Case': 'Par', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Gender=Fem': {POS: PROPN, 'Animacy': 'Inan', 'Gender': 'Fem'},
|
||||
'PROPN__Animacy=Inan|Gender=Masc': {POS: PROPN, 'Animacy': 'Inan', 'Gender': 'Masc'},
|
||||
'PROPN__Animacy=Inan|Gender=Masc|Number=Plur': {POS: PROPN, 'Animacy': 'Inan', 'Gender': 'Masc', 'Number': 'Plur'},
|
||||
'PROPN__Animacy=Inan|Gender=Masc|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Animacy=Inan|Gender=Neut|Number=Sing': {POS: PROPN, 'Animacy': 'Inan', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PROPN__Case=Acc|Degree=Pos|Gender=Fem|Number=Sing': {POS: PROPN, 'Case': 'Acc', 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Case=Dat|Degree=Pos|Gender=Masc|Number=Sing': {POS: PROPN, 'Case': 'Dat', 'Degree': 'Pos', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Case=Ins|Degree=Pos|Gender=Fem|Number=Sing': {POS: PROPN, 'Case': 'Ins', 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Case=Ins|Degree=Pos|Number=Plur': {POS: PROPN, 'Case': 'Ins', 'Degree': 'Pos', 'Number': 'Plur'},
|
||||
'PROPN__Case=Nom|Degree=Pos|Gender=Fem|Number=Sing': {POS: PROPN, 'Case': 'Nom', 'Degree': 'Pos', 'Gender': 'Fem', 'Number': 'Sing'},
|
||||
'PROPN__Case=Nom|Degree=Pos|Gender=Masc|Number=Sing': {POS: PROPN, 'Case': 'Nom', 'Degree': 'Pos', 'Gender': 'Masc', 'Number': 'Sing'},
|
||||
'PROPN__Case=Nom|Degree=Pos|Gender=Neut|Number=Sing': {POS: PROPN, 'Case': 'Nom', 'Degree': 'Pos', 'Gender': 'Neut', 'Number': 'Sing'},
|
||||
'PROPN__Case=Nom|Degree=Pos|Number=Plur': {POS: PROPN, 'Case': 'Nom', 'Degree': 'Pos', 'Number': 'Plur'},
|
||||
'PROPN__Degree=Pos|Gender=Neut|Number=Sing|Variant=Short': {POS: PROPN, 'Degree': 'Pos', 'Gender': 'Neut', 'Number': 'Sing', 'Variant': 'Short'},
|
||||
'PROPN__Degree=Pos|Number=Plur|Variant=Short': {POS: PROPN, 'Degree': 'Pos', 'Number': 'Plur', 'Variant': 'Short'},
|
||||
'PROPN__Foreign=Yes': {POS: PROPN, 'Foreign': 'Yes'},
|
||||
'PROPN__Number=Sing': {POS: PROPN, 'Number': 'Sing'},
|
||||
'PROPN___': {POS: PROPN},
|
||||
'PROPN': {POS: PROPN},
|
||||
'PUNCT___': {POS: PUNCT},
|
||||
'PUNCT': {POS: PUNCT},
|
||||
'SCONJ__Mood=Cnd': {POS: SCONJ, 'Mood': 'Cnd'},
|
||||
'SCONJ___': {POS: SCONJ},
|
||||
'SCONJ': {POS: SCONJ},
|
||||
'SYM___': {POS: SYM},
|
||||
'SYM': {POS: SYM},
|
||||
'VERB__Animacy=Anim|Aspect=Imp|Case=Acc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Animacy=Anim|Aspect=Imp|Case=Acc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Animacy=Anim|Aspect=Imp|Case=Acc|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Animacy=Anim|Aspect=Imp|Case=Acc|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Animacy=Anim|Aspect=Imp|Case=Acc|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Animacy=Anim|Aspect=Imp|Case=Acc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Imp', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Animacy=Anim|Aspect=Imp|Case=Acc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Imp', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Animacy=Anim|Aspect=Imp|Case=Acc|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Imp', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Animacy=Anim|Aspect=Imp|Case=Acc|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Imp', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Animacy=Anim|Aspect=Imp|Case=Acc|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Imp', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Animacy=Anim|Aspect=Perf|Case=Acc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Perf', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Animacy=Anim|Aspect=Perf|Case=Acc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Perf', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Animacy=Anim|Aspect=Perf|Case=Acc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Perf', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Animacy=Anim|Aspect=Perf|Case=Acc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Perf', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Animacy=Anim|Aspect=Perf|Case=Acc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Perf', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Animacy=Anim|Aspect=Perf|Case=Acc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Animacy': 'Anim', 'Aspect': 'Perf', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Animacy=Inan|Aspect=Imp|Case=Acc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Animacy=Inan|Aspect=Imp|Case=Acc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Animacy=Inan|Aspect=Imp|Case=Acc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Animacy=Inan|Aspect=Imp|Case=Acc|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Animacy=Inan|Aspect=Imp|Case=Acc|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Animacy=Inan|Aspect=Imp|Case=Acc|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Animacy=Inan|Aspect=Imp|Case=Acc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Imp', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Animacy=Inan|Aspect=Imp|Case=Acc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Imp', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Animacy=Inan|Aspect=Imp|Case=Acc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Imp', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Animacy=Inan|Aspect=Imp|Case=Acc|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Imp', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Animacy=Inan|Aspect=Imp|Case=Acc|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Imp', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Animacy=Inan|Aspect=Imp|Case=Acc|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Imp', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Animacy=Inan|Aspect=Perf|Case=Acc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Perf', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Animacy=Inan|Aspect=Perf|Case=Acc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Perf', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Animacy=Inan|Aspect=Perf|Case=Acc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Perf', 'Case': 'Acc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Animacy=Inan|Aspect=Perf|Case=Acc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Perf', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Animacy=Inan|Aspect=Perf|Case=Acc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Perf', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Animacy=Inan|Aspect=Perf|Case=Acc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Animacy': 'Inan', 'Aspect': 'Perf', 'Case': 'Acc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Acc|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Acc|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Acc|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Acc|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Acc|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Acc|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Acc|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Acc|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Acc|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Acc|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Acc|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Acc|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Dat|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Dat', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Gen|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Gen', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Ins|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Ins', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Loc|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Loc', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Fem|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Masc|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Gender=Neut|Number=Sing|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Case=Nom|Number=Plur|Tense=Pres|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Case': 'Nom', 'Number': 'Plur', 'Tense': 'Pres', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Gender=Fem|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Fem', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Gender=Fem|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Fem', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Gender=Fem|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Fem', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Gender=Fem|Number=Sing|Tense=Past|Variant=Short|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'Variant': 'Short', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Gender=Fem|Number=Sing|Tense=Pres|Variant=Short|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Pres', 'Variant': 'Short', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Gender=Masc|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Masc', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Gender=Masc|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Masc', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Gender=Masc|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Masc', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Gender=Masc|Number=Sing|Tense=Past|Variant=Short|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'Variant': 'Short', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Gender=Masc|Number=Sing|Tense=Pres|Variant=Short|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Pres', 'Variant': 'Short', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Gender=Neut|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Neut', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Gender=Neut|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Neut', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Gender=Neut|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Neut', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Gender=Neut|Number=Sing|Tense=Past|Variant=Short|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'Variant': 'Short', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Gender=Neut|Number=Sing|Tense=Pres|Variant=Short|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Pres', 'Variant': 'Short', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Mood=Imp|Number=Plur|Person=2|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Imp', 'Number': 'Plur', 'Person': '2', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Mood=Imp|Number=Plur|Person=2|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Imp', 'Number': 'Plur', 'Person': '2', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Mood=Imp|Number=Sing|Person=2|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Imp', 'Number': 'Sing', 'Person': '2', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Mood=Imp|Number=Sing|Person=2|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Imp', 'Number': 'Sing', 'Person': '2', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '1', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Plur|Person=1|Tense=Pres|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '1', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '2', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Plur|Person=2|Tense=Pres|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '2', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '3', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '3', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Plur|Person=3|Tense=Pres|VerbForm=Fin|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '3', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Plur|Tense=Past|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Plur|Tense=Past|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Plur|Tense=Past|VerbForm=Fin|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '1', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Sing|Person=1|Tense=Pres|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '1', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '2', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Sing|Person=2|Tense=Pres|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '2', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '3', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '3', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '3', 'Tense': 'Pres', 'VerbForm': 'Fin', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Number=Plur|Tense=Past|Variant=Short|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Number': 'Plur', 'Tense': 'Past', 'Variant': 'Short', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Number=Plur|Tense=Pres|Variant=Short|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Number': 'Plur', 'Tense': 'Pres', 'Variant': 'Short', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|Tense=Past|VerbForm=Conv|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Tense': 'Past', 'VerbForm': 'Conv', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Tense=Pres|VerbForm=Conv|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'Tense': 'Pres', 'VerbForm': 'Conv', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|Tense=Pres|VerbForm=Conv|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'Tense': 'Pres', 'VerbForm': 'Conv', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|Tense=Pres|VerbForm=Conv|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'Tense': 'Pres', 'VerbForm': 'Conv', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Imp|VerbForm=Inf|Voice=Act': {POS: VERB, 'Aspect': 'Imp', 'VerbForm': 'Inf', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Imp|VerbForm=Inf|Voice=Mid': {POS: VERB, 'Aspect': 'Imp', 'VerbForm': 'Inf', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Imp|VerbForm=Inf|Voice=Pass': {POS: VERB, 'Aspect': 'Imp', 'VerbForm': 'Inf', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Acc|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Acc|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Acc|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Acc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Acc|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Acc|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Acc|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Acc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Dat|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Dat|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Dat|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Dat', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Dat|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Dat|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Dat|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Dat', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Dat|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Dat|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Dat|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Dat', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Dat|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Dat', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Dat|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Dat', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Dat|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Dat', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Number=Plur|Tense=Fut|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Number': 'Plur', 'Tense': 'Fut', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Gen|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Gen', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Ins|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Ins|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Ins|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Ins', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Ins|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Ins|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Ins|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Ins', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Ins|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Ins|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Ins|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Ins', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Ins|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Ins', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Ins|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Ins', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Ins|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Ins', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Loc|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Loc|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Loc|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Loc', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Loc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Loc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Loc|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Loc', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Loc|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Loc|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Loc|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Loc', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Loc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Loc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Loc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Loc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Loc|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Loc', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Nom|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Nom|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Nom|Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Nom', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Nom|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Nom|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Nom|Gender=Masc|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Nom', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Nom|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Nom|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Nom|Gender=Neut|Number=Sing|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Nom', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Case=Nom|Number=Plur|Tense=Past|VerbForm=Part|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Nom', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Case=Nom|Number=Plur|Tense=Past|VerbForm=Part|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Nom', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Case=Nom|Number=Plur|Tense=Past|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Case': 'Nom', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Gender=Fem|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Gender': 'Fem', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Gender=Fem|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Gender': 'Fem', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Gender=Fem|Number=Sing|Tense=Past|Variant=Short|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Gender': 'Fem', 'Number': 'Sing', 'Tense': 'Past', 'Variant': 'Short', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Gender=Masc|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Gender': 'Masc', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Gender=Masc|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Gender': 'Masc', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Gender=Masc|Number=Sing|Tense=Past|Variant=Short|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Gender': 'Masc', 'Number': 'Sing', 'Tense': 'Past', 'Variant': 'Short', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Gender=Neut|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Gender': 'Neut', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Gender=Neut|Mood=Ind|Number=Sing|Tense=Past|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Gender': 'Neut', 'Mood': 'Ind', 'Number': 'Sing', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Gender=Neut|Number=Sing|Tense=Past|Variant=Short|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Gender': 'Neut', 'Number': 'Sing', 'Tense': 'Past', 'Variant': 'Short', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Mood=Imp|Number=Plur|Person=1|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Imp', 'Number': 'Plur', 'Person': '1', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Mood=Imp|Number=Plur|Person=2|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Imp', 'Number': 'Plur', 'Person': '2', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Mood=Imp|Number=Plur|Person=2|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Imp', 'Number': 'Plur', 'Person': '2', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Mood=Imp|Number=Sing|Person=2|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Imp', 'Number': 'Sing', 'Person': '2', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Mood=Imp|Number=Sing|Person=2|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Imp', 'Number': 'Sing', 'Person': '2', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '1', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Plur|Person=1|Tense=Fut|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '1', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Plur|Person=2|Tense=Fut|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '2', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Plur|Person=2|Tense=Fut|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '2', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '3', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '3', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Plur|Person=3|Tense=Fut|VerbForm=Fin|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Plur', 'Person': '3', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Plur|Tense=Past|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Plur|Tense=Past|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Plur', 'Tense': 'Past', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Sing|Person=1|Tense=Fut|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '1', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Sing|Person=1|Tense=Fut|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '1', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Sing|Person=2|Tense=Fut|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '2', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Sing|Person=2|Tense=Fut|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '2', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '3', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Mood=Ind|Number=Sing|Person=3|Tense=Fut|VerbForm=Fin|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Mood': 'Ind', 'Number': 'Sing', 'Person': '3', 'Tense': 'Fut', 'VerbForm': 'Fin', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|Number=Plur|Tense=Past|Variant=Short|VerbForm=Part|Voice=Pass': {POS: VERB, 'Aspect': 'Perf', 'Number': 'Plur', 'Tense': 'Past', 'Variant': 'Short', 'VerbForm': 'Part', 'Voice': 'Pass'},
|
||||
'VERB__Aspect=Perf|Tense=Past|VerbForm=Conv|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'Tense': 'Past', 'VerbForm': 'Conv', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|Tense=Past|VerbForm=Conv|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'Tense': 'Past', 'VerbForm': 'Conv', 'Voice': 'Mid'},
|
||||
'VERB__Aspect=Perf|VerbForm=Inf|Voice=Act': {POS: VERB, 'Aspect': 'Perf', 'VerbForm': 'Inf', 'Voice': 'Act'},
|
||||
'VERB__Aspect=Perf|VerbForm=Inf|Voice=Mid': {POS: VERB, 'Aspect': 'Perf', 'VerbForm': 'Inf', 'Voice': 'Mid'},
|
||||
'VERB__Voice=Act': {POS: VERB, 'Voice': 'Act'},
|
||||
'VERB___': {POS: VERB},
|
||||
'VERB': {POS: VERB},
|
||||
'X__Foreign=Yes': {POS: X, 'Foreign': 'Yes'},
|
||||
'X___': {POS: X},
|
||||
'X': {POS: X},
|
||||
}
|
||||
# fmt: on
|
||||
|
|
@ -1,5 +1,4 @@
|
|||
from .stop_words import STOP_WORDS
|
||||
from .tag_map import TAG_MAP
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
|
||||
from ...language import Language
|
||||
|
|
@ -10,7 +9,6 @@ class SlovakDefaults(Language.Defaults):
|
|||
lex_attr_getters = dict(Language.Defaults.lex_attr_getters)
|
||||
lex_attr_getters.update(LEX_ATTRS)
|
||||
lex_attr_getters[LANG] = lambda text: "sk"
|
||||
tag_map = TAG_MAP
|
||||
stop_words = STOP_WORDS
|
||||
|
||||
|
||||
|
|
|
|||
File diff suppressed because it is too large
Load Diff
|
|
@ -1,8 +1,6 @@
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .tag_map import TAG_MAP
|
||||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .morph_rules import MORPH_RULES
|
||||
|
||||
# Punctuation stolen from Danish
|
||||
from ..da.punctuation import TOKENIZER_INFIXES, TOKENIZER_SUFFIXES
|
||||
|
|
@ -23,12 +21,9 @@ class SwedishDefaults(Language.Defaults):
|
|||
Language.Defaults.lex_attr_getters[NORM], BASE_NORMS
|
||||
)
|
||||
tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS)
|
||||
morph_rules = MORPH_RULES
|
||||
tag_map = TAG_MAP
|
||||
infixes = TOKENIZER_INFIXES
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
stop_words = STOP_WORDS
|
||||
morph_rules = MORPH_RULES
|
||||
syntax_iterators = SYNTAX_ITERATORS
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -1,285 +0,0 @@
|
|||
from ...symbols import LEMMA, PRON_LEMMA
|
||||
|
||||
|
||||
# Used the table of pronouns at https://sv.wiktionary.org/wiki/deras
|
||||
MORPH_RULES = {
|
||||
"PRP": {
|
||||
"jag": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"mig": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"mej": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"du": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"han": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"honom": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Masc",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"hon": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Fem",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"henne": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Fem",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"det": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Sing",
|
||||
"Gender": "Neut",
|
||||
},
|
||||
"vi": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"oss": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"ni": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Plur",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"er": {LEMMA: PRON_LEMMA, "PronType": "Prs", "Person": "Two", "Number": "Plur"},
|
||||
"de": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Case": "Nom",
|
||||
},
|
||||
"dom": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Case": "Nom,Acc",
|
||||
},
|
||||
"dem": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Case": "Acc",
|
||||
},
|
||||
"min": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"mitt": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"mina": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"din": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"ditt": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"dina": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"hans": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing,Plur",
|
||||
"Gender": "Masc",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"hennes": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing,Plur",
|
||||
"Gender": "Fem",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"dess": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Sing,Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"vår": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"våran": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"vårt": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"vårat": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"våra": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "One",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"eran": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"ert": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"erat": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"era": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Two",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
"deras": {
|
||||
LEMMA: PRON_LEMMA,
|
||||
"PronType": "Prs",
|
||||
"Person": "Three",
|
||||
"Number": "Plur",
|
||||
"Poss": "Yes",
|
||||
"Reflex": "Yes",
|
||||
},
|
||||
},
|
||||
"VBZ": {
|
||||
"är": {
|
||||
"VerbForm": "Fin",
|
||||
"Person": "One,Two,Three",
|
||||
"Tense": "Pres",
|
||||
"Mood": "Ind",
|
||||
}
|
||||
},
|
||||
"VBP": {"är": {"VerbForm": "Fin", "Tense": "Pres", "Mood": "Ind"}},
|
||||
"VBD": {
|
||||
"var": {"VerbForm": "Fin", "Tense": "Past", "Number": "Sing"},
|
||||
"vart": {"VerbForm": "Fin", "Tense": "Past", "Number": "Plur"},
|
||||
},
|
||||
}
|
||||
|
|
@ -1,188 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, ADJ, CCONJ, SCONJ, NUM, DET, ADV
|
||||
from ...symbols import ADP, X, VERB, NOUN, PROPN, PART, INTJ, PRON
|
||||
|
||||
|
||||
# Tag mappings according to https://universaldependencies.org/tagset-conversion/sv-suc-uposf.html
|
||||
# for https://github.com/UniversalDependencies/UD_Swedish-Talbanken
|
||||
|
||||
TAG_MAP = {
|
||||
"AB": {POS: ADV}, # inte, också, så, bara, nu
|
||||
"AB|AN": {POS: ADV}, # t.ex., ca, t_ex, bl.a., s_k
|
||||
"AB|KOM": {POS: ADV}, # mer, tidigare, mindre, vidare, mera
|
||||
"AB|POS": {POS: ADV}, # mycket, helt, ofta, länge, långt
|
||||
"AB|SMS": {POS: ADV}, # över-, in-
|
||||
"AB|SUV": {POS: ADV}, # minst, mest, högst, främst, helst
|
||||
"DT|MAS|SIN|DEF": {POS: DET},
|
||||
"DT|MAS|SIN|IND": {POS: DET},
|
||||
"DT|NEU|SIN|DEF": {POS: DET}, # det, detta
|
||||
"DT|NEU|SIN|IND": {POS: DET}, # ett, något, inget, vart, vartannat
|
||||
"DT|NEU|SIN|IND/DEF": {POS: DET}, # allt
|
||||
"DT|UTR/NEU|PLU|DEF": {POS: DET}, # de, dessa, bägge, dom
|
||||
"DT|UTR/NEU|PLU|IND": {POS: DET}, # några, inga
|
||||
"DT|UTR/NEU|PLU|IND/DEF": {POS: DET}, # alla
|
||||
"DT|UTR/NEU|SIN/PLU|IND": {POS: DET}, # samma
|
||||
"DT|UTR/NEU|SIN|DEF": {POS: DET}, # vardera
|
||||
"DT|UTR/NEU|SIN|IND": {POS: DET}, # varje, varenda
|
||||
"DT|UTR|SIN|DEF": {POS: DET}, # den, denna
|
||||
"DT|UTR|SIN|IND": {POS: DET}, # en, någon, ingen, var, varannan
|
||||
"DT|UTR|SIN|IND/DEF": {POS: DET}, # all
|
||||
"HA": {POS: ADV}, # när, där, hur, som, då
|
||||
"HD|NEU|SIN|IND": {POS: DET}, # vilket
|
||||
"HD|UTR/NEU|PLU|IND": {POS: DET}, # vilka
|
||||
"HD|UTR|SIN|IND": {POS: DET}, # vilken
|
||||
"HP|-|-|-": {POS: PRON}, # som
|
||||
"HP|NEU|SIN|IND": {POS: PRON}, # vad, vilket
|
||||
"HP|NEU|SIN|IND|SMS": {POS: PRON},
|
||||
"HP|UTR/NEU|PLU|IND": {POS: PRON}, # vilka
|
||||
"HP|UTR|SIN|IND": {POS: PRON}, # vilken, vem
|
||||
"HS|DEF": {POS: DET}, # vars, vilkas, Vems
|
||||
"IE": {POS: PART}, # att
|
||||
"IN": {POS: INTJ}, # Jo, ja, nej, fan, visst
|
||||
"JJ|AN": {POS: ADJ}, # ev, S:t, Kungl, Kungl., Teol
|
||||
"JJ|KOM|UTR/NEU|SIN/PLU|IND/DEF|GEN": {POS: ADJ}, # äldres
|
||||
"JJ|KOM|UTR/NEU|SIN/PLU|IND/DEF|NOM": {
|
||||
POS: ADJ
|
||||
}, # större, högre, mindre, bättre, äldre
|
||||
"JJ|KOM|UTR/NEU|SIN/PLU|IND/DEF|SMS": {POS: ADJ},
|
||||
"JJ|POS|MAS|SIN|DEF|GEN": {POS: ADJ}, # enskildes, sjukes, andres
|
||||
"JJ|POS|MAS|SIN|DEF|NOM": {POS: ADJ}, # enskilde, sjuke, andre, unge, ene
|
||||
"JJ|POS|NEU|SIN|IND/DEF|NOM": {POS: ADJ}, # eget
|
||||
"JJ|POS|NEU|SIN|IND|GEN": {POS: ADJ},
|
||||
"JJ|POS|NEU|SIN|IND|NOM": {POS: ADJ}, # annat, svårt, möjligt, nytt, sådant
|
||||
"JJ|POS|UTR/NEU|PLU|IND/DEF|GEN": {
|
||||
POS: ADJ
|
||||
}, # ogiftas, ungas, frånskildas, efterkommandes, färgblindas
|
||||
"JJ|POS|UTR/NEU|PLU|IND/DEF|NOM": {POS: ADJ}, # olika, andra, många, stora, vissa
|
||||
"JJ|POS|UTR/NEU|PLU|IND|NOM": {POS: ADJ}, # flera, sådana, fler, få, samtliga
|
||||
"JJ|POS|UTR/NEU|SIN/PLU|IND|NOM": {POS: ADJ},
|
||||
"JJ|POS|UTR/NEU|SIN/PLU|IND/DEF|NOM": {POS: ADJ}, # bra, ena, enda, nästa, ringa
|
||||
"JJ|POS|UTR/NEU|SIN|DEF|GEN": {POS: ADJ},
|
||||
"JJ|POS|UTR/NEU|SIN|DEF|NOM": {POS: ADJ}, # hela, nya, andra, svenska, ekonomiska
|
||||
"JJ|POS|UTR|-|-|SMS": {POS: ADJ}, # fri-, låg-, sexual-
|
||||
"JJ|POS|UTR|SIN|IND/DEF|NOM": {POS: ADJ}, # egen
|
||||
"JJ|POS|UTR|SIN|IND|GEN": {POS: ADJ}, # enskilds
|
||||
"JJ|POS|UTR|SIN|IND|NOM": {POS: ADJ}, # stor, annan, själv, sådan, viss
|
||||
"JJ|SUV|MAS|SIN|DEF|GEN": {POS: ADJ},
|
||||
"JJ|SUV|MAS|SIN|DEF|NOM": {POS: ADJ}, # störste, främste, äldste, minste
|
||||
"JJ|SUV|UTR/NEU|PLU|DEF|NOM": {POS: ADJ}, # flesta
|
||||
"JJ|SUV|UTR/NEU|PLU|IND|NOM": {POS: ADJ},
|
||||
"JJ|SUV|UTR/NEU|SIN/PLU|DEF|NOM": {
|
||||
POS: ADJ
|
||||
}, # bästa, största, närmaste, viktigaste, högsta
|
||||
"JJ|SUV|UTR/NEU|SIN/PLU|IND|NOM": {
|
||||
POS: ADJ
|
||||
}, # störst, bäst, tidigast, högst, fattigast
|
||||
"KN": {POS: CCONJ}, # och, eller, som, än, men
|
||||
"KN|AN": {POS: CCONJ},
|
||||
"MAD": {POS: PUNCT}, # ., ?, :, !, ...
|
||||
"MID": {POS: PUNCT}, # ,, -, :, *, ;
|
||||
"NN|-|-|-|-": {POS: NOUN}, # godo, fjol, fullo, somras, måtto
|
||||
"NN|AN": {POS: NOUN}, # kr, %, s., dr, kap.
|
||||
"NN|NEU|-|-|-": {POS: NOUN},
|
||||
"NN|NEU|-|-|SMS": {POS: NOUN}, # yrkes-, barn-, hem-, fack-, vatten-
|
||||
"NN|NEU|PLU|DEF|GEN": {
|
||||
POS: NOUN
|
||||
}, # barnens, årens, u-ländernas, företagens, århundradenas
|
||||
"NN|NEU|PLU|DEF|NOM": {POS: NOUN}, # barnen, u-länderna, åren, länderna, könen
|
||||
"NN|NEU|PLU|IND|GEN": {POS: NOUN}, # slags, års, barns, länders, tusentals
|
||||
"NN|NEU|PLU|IND|NOM": {POS: NOUN}, # barn, år, fall, länder, problem
|
||||
"NN|NEU|SIN|DEF|GEN": {
|
||||
POS: NOUN
|
||||
}, # äktenskapets, samhällets, barnets, 1800-talets, 1960-talets
|
||||
"NN|NEU|SIN|DEF|NOM": {
|
||||
POS: NOUN
|
||||
}, # äktenskapet, samhället, barnet, stället, hemmet
|
||||
"NN|NEU|SIN|IND|GEN": {POS: NOUN}, # års, slags, lands, havs, företags
|
||||
"NN|NEU|SIN|IND|NOM": {POS: NOUN}, # år, arbete, barn, sätt, äktenskap
|
||||
"NN|SMS": {POS: NOUN}, # PCB-, Syd-
|
||||
"NN|UTR|-|-|-": {POS: NOUN}, # dags, rätta
|
||||
"NN|UTR|-|-|SMS": {POS: NOUN}, # far-, kibbutz-, röntgen-, barna-, hälso-
|
||||
"NN|UTR|PLU|DEF|GEN": {
|
||||
POS: NOUN
|
||||
}, # föräldrarnas, kvinnornas, elevernas, kibbutzernas, makarnas
|
||||
"NN|UTR|PLU|DEF|NOM": {
|
||||
POS: NOUN
|
||||
}, # kvinnorna, föräldrarna, makarna, männen, hyrorna
|
||||
"NN|UTR|PLU|IND|GEN": {POS: NOUN}, # människors, kvinnors, dagars, tiders, månaders
|
||||
"NN|UTR|PLU|IND|NOM": {POS: NOUN}, # procent, människor, kvinnor, miljoner, kronor
|
||||
"NN|UTR|SIN|DEF|GEN": {POS: NOUN}, # kvinnans, världens, familjens, dagens, jordens
|
||||
"NN|UTR|SIN|DEF|NOM": {POS: NOUN}, # familjen, kvinnan, mannen, världen, skolan
|
||||
"NN|UTR|SIN|IND|GEN": {POS: NOUN}, # sorts, medelålders, makes, kvinnas, veckas
|
||||
"NN|UTR|SIN|IND|NOM": {POS: NOUN}, # del, tid, dag, fråga, man
|
||||
"PAD": {POS: PUNCT}, # , ), (
|
||||
"PC|AN": {POS: VERB},
|
||||
"PC|PRF|MAS|SIN|DEF|GEN": {POS: VERB}, # avlidnes
|
||||
"PC|PRF|MAS|SIN|DEF|NOM": {POS: VERB},
|
||||
"PC|PRF|NEU|SIN|IND|NOM": {POS: VERB}, # taget, sett, särskilt, förbjudet, ökat
|
||||
"PC|PRF|UTR/NEU|PLU|IND/DEF|GEN": {POS: VERB}, # försäkrades, anställdas
|
||||
"PC|PRF|UTR/NEU|PLU|IND/DEF|NOM": {
|
||||
POS: VERB
|
||||
}, # särskilda, gifta, ökade, handikappade, skilda
|
||||
"PC|PRF|UTR/NEU|SIN|DEF|GEN": {POS: VERB},
|
||||
"PC|PRF|UTR/NEU|SIN|DEF|NOM": {POS: VERB}, # ökade, gifta, nämnda, nedärvda, dolda
|
||||
"PC|PRF|UTR|SIN|IND|GEN": {POS: VERB},
|
||||
"PC|PRF|UTR|SIN|IND|NOM": {POS: VERB}, # särskild, ökad, beredd, gift, oförändrad
|
||||
"PC|PRS|UTR/NEU|SIN/PLU|IND/DEF|GEN": {
|
||||
POS: VERB
|
||||
}, # studerandes, sammanboendes, dubbelarbetandes
|
||||
"PC|PRS|UTR/NEU|SIN/PLU|IND/DEF|NOM": {
|
||||
POS: VERB
|
||||
}, # följande, beroende, nuvarande, motsvarande, liknande
|
||||
"PL": {POS: PART}, # ut, upp, in, till, med
|
||||
"PL|SMS": {POS: PART},
|
||||
"PM": {POS: PROPN}, # F, N, Liechtenstein, Danmark, DK
|
||||
"PM|GEN": {POS: PROPN}, # Sveriges, EEC:s, Guds, Stockholms, Kristi
|
||||
"PM|NOM": {POS: PROPN}, # Sverige, EEC, Stockholm, USA, ATP
|
||||
"PM|SMS": {POS: PROPN}, # Göteborgs-, Nord-, Väst-
|
||||
"PN|MAS|SIN|DEF|SUB/OBJ": {POS: PRON}, # denne
|
||||
"PN|NEU|SIN|DEF|SUB/OBJ": {POS: PRON}, # det, detta, detsamma
|
||||
"PN|NEU|SIN|IND|SUB/OBJ": {POS: PRON}, # något, allt, mycket, annat, ingenting
|
||||
"PN|UTR/NEU|PLU|DEF|OBJ": {POS: PRON}, # dem, varandra, varann
|
||||
"PN|UTR/NEU|PLU|DEF|SUB": {POS: PRON}, # de, bägge
|
||||
"PN|UTR/NEU|PLU|DEF|SUB/OBJ": {POS: PRON}, # dessa, dom, båda, den, bådadera
|
||||
"PN|UTR/NEU|PLU|IND|SUB/OBJ": {POS: PRON}, # andra, alla, många, sådana, några
|
||||
"PN|UTR/NEU|SIN/PLU|DEF|OBJ": {POS: PRON}, # sig, sej
|
||||
"PN|UTR|PLU|DEF|OBJ": {POS: PRON}, # oss, er, eder
|
||||
"PN|UTR|PLU|DEF|SUB": {POS: PRON}, # vi
|
||||
"PN|UTR|SIN|DEF|OBJ": {POS: PRON}, # dig, mig, henne, honom, Er
|
||||
"PN|UTR|SIN|DEF|SUB": {POS: PRON}, # du, han, hon, jag, ni
|
||||
"PN|UTR|SIN|DEF|SUB/OBJ": {POS: PRON}, # den, denna, densamma
|
||||
"PN|UTR|SIN|IND|SUB": {POS: PRON}, # man
|
||||
"PN|UTR|SIN|IND|SUB/OBJ": {POS: PRON}, # en, var, någon, ingen, Varannan
|
||||
"PP": {POS: ADP}, # i, av, på, för, till
|
||||
"PP|AN": {POS: ADP}, # f
|
||||
"PS|AN": {POS: DET},
|
||||
"PS|NEU|SIN|DEF": {POS: DET}, # sitt, vårt, ditt, mitt, ert
|
||||
"PS|UTR/NEU|PLU|DEF": {POS: DET}, # sina, våra, dina, mina
|
||||
"PS|UTR/NEU|SIN/PLU|DEF": {POS: DET}, # deras, dess, hans, hennes, varandras
|
||||
"PS|UTR|SIN|DEF": {POS: DET}, # sin, vår, din, min, er
|
||||
"RG": {POS: NUM}, # 2, 17, 20, 1, 18
|
||||
"RG|GEN": {POS: NUM},
|
||||
"RG|MAS|SIN|DEF|NOM": {POS: NUM},
|
||||
"RG|NEU|SIN|IND|NOM": {POS: NUM}, # ett
|
||||
"RG|NOM": {POS: NUM}, # två, tre, 1, 20, 2
|
||||
"RG|SMS": {POS: NUM}, # ett-, 1950-, två-, tre-, 1700-
|
||||
"RG|UTR/NEU|SIN|DEF|NOM": {POS: NUM},
|
||||
"RG|UTR|SIN|IND|NOM": {POS: NUM}, # en
|
||||
"RO|MAS|SIN|IND/DEF|GEN": {POS: ADJ},
|
||||
"RO|MAS|SIN|IND/DEF|NOM": {POS: ADJ}, # förste
|
||||
"RO|GEN": {POS: ADJ},
|
||||
"RO|NOM": {POS: ADJ}, # första, andra, tredje, fjärde, femte
|
||||
"SN": {POS: SCONJ}, # att, om, innan, eftersom, medan
|
||||
"UO": {POS: X}, # companionship, vice, versa, family, capita
|
||||
"VB|AN": {POS: VERB}, # jfr
|
||||
"VB|IMP|AKT": {POS: VERB}, # se, Diskutera, låt, Läs, Gå
|
||||
"VB|IMP|SFO": {POS: VERB}, # tas
|
||||
"VB|INF|AKT": {POS: VERB}, # vara, få, ha, bli, kunna
|
||||
"VB|INF|SFO": {POS: VERB}, # användas, finnas, göras, tas, ses
|
||||
"VB|KON|PRS|AKT": {POS: VERB}, # vare, Gånge
|
||||
"VB|KON|PRT|AKT": {POS: VERB}, # vore, finge
|
||||
"VB|KON|PRT|SFO": {POS: VERB},
|
||||
"VB|PRS|AKT": {POS: VERB}, # är, har, kan, får, måste
|
||||
"VB|PRS|SFO": {POS: VERB}, # finns, kallas, behövs, beräknas, används
|
||||
"VB|PRT|AKT": {POS: VERB}, # skulle, var, hade, kunde, fick
|
||||
"VB|PRT|SFO": {POS: VERB}, # fanns, gjordes, höjdes, användes, infördes
|
||||
"VB|SMS": {POS: VERB}, # läs-
|
||||
"VB|SUP|AKT": {POS: VERB}, # varit, fått, blivit, haft, kommit
|
||||
"VB|SUP|SFO": {POS: VERB}, # nämnts, gjorts, förändrats, sagts, framhållits
|
||||
}
|
||||
|
|
@ -1,5 +1,4 @@
|
|||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS
|
||||
from .tag_map import TAG_MAP
|
||||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
|
||||
|
|
@ -33,7 +32,6 @@ class ThaiDefaults(Language.Defaults):
|
|||
lex_attr_getters.update(LEX_ATTRS)
|
||||
lex_attr_getters[LANG] = lambda _text: "th"
|
||||
tokenizer_exceptions = dict(TOKENIZER_EXCEPTIONS)
|
||||
tag_map = TAG_MAP
|
||||
stop_words = STOP_WORDS
|
||||
|
||||
@classmethod
|
||||
|
|
|
|||
|
|
@ -1,86 +0,0 @@
|
|||
from ...symbols import POS, NOUN, PRON, ADJ, ADV, INTJ, PROPN, DET, NUM, AUX, VERB
|
||||
from ...symbols import ADP, CCONJ, PART, PUNCT, SPACE, SCONJ
|
||||
|
||||
# Source: Korakot Chaovavanich
|
||||
# https://www.facebook.com/photo.php?fbid=390564854695031&set=p.390564854695031&type=3&permPage=1&ifg=1
|
||||
TAG_MAP = {
|
||||
# NOUN
|
||||
"NOUN": {POS: NOUN},
|
||||
"NCMN": {POS: NOUN},
|
||||
"NTTL": {POS: NOUN},
|
||||
"CNIT": {POS: NOUN},
|
||||
"CLTV": {POS: NOUN},
|
||||
"CMTR": {POS: NOUN},
|
||||
"CFQC": {POS: NOUN},
|
||||
"CVBL": {POS: NOUN},
|
||||
# VERB
|
||||
"VACT": {POS: VERB},
|
||||
"VSTA": {POS: VERB},
|
||||
# PRON
|
||||
"PRON": {POS: PRON},
|
||||
"NPRP": {POS: PRON},
|
||||
# ADJ
|
||||
"ADJ": {POS: ADJ},
|
||||
"NONM": {POS: ADJ},
|
||||
"VATT": {POS: ADJ},
|
||||
"DONM": {POS: ADJ},
|
||||
# ADV
|
||||
"ADV": {POS: ADV},
|
||||
"ADVN": {POS: ADV},
|
||||
"ADVI": {POS: ADV},
|
||||
"ADVP": {POS: ADV},
|
||||
"ADVS": {POS: ADV},
|
||||
# INT
|
||||
"INT": {POS: INTJ},
|
||||
# PRON
|
||||
"PROPN": {POS: PROPN},
|
||||
"PPRS": {POS: PROPN},
|
||||
"PDMN": {POS: PROPN},
|
||||
"PNTR": {POS: PROPN},
|
||||
# DET
|
||||
"DET": {POS: DET},
|
||||
"DDAN": {POS: DET},
|
||||
"DDAC": {POS: DET},
|
||||
"DDBQ": {POS: DET},
|
||||
"DDAQ": {POS: DET},
|
||||
"DIAC": {POS: DET},
|
||||
"DIBQ": {POS: DET},
|
||||
"DIAQ": {POS: DET},
|
||||
# TODO: resolve duplicate (see below)
|
||||
# "DCNM": {POS: DET},
|
||||
# NUM
|
||||
"NUM": {POS: NUM},
|
||||
"NCNM": {POS: NUM},
|
||||
"NLBL": {POS: NUM},
|
||||
"DCNM": {POS: NUM},
|
||||
# AUX
|
||||
"AUX": {POS: AUX},
|
||||
"XVBM": {POS: AUX},
|
||||
"XVAM": {POS: AUX},
|
||||
"XVMM": {POS: AUX},
|
||||
"XVBB": {POS: AUX},
|
||||
"XVAE": {POS: AUX},
|
||||
# ADP
|
||||
"ADP": {POS: ADP},
|
||||
"RPRE": {POS: ADP},
|
||||
# CCONJ
|
||||
"CCONJ": {POS: CCONJ},
|
||||
"JCRG": {POS: CCONJ},
|
||||
# SCONJ
|
||||
"SCONJ": {POS: SCONJ},
|
||||
"PREL": {POS: SCONJ},
|
||||
"JSBR": {POS: SCONJ},
|
||||
"JCMP": {POS: SCONJ},
|
||||
# PART
|
||||
"PART": {POS: PART},
|
||||
"FIXN": {POS: PART},
|
||||
"FIXV": {POS: PART},
|
||||
"EAFF": {POS: PART},
|
||||
"AITT": {POS: PART},
|
||||
"NEG": {POS: PART},
|
||||
"EITT": {POS: PART},
|
||||
# PUNCT
|
||||
"PUNCT": {POS: PUNCT},
|
||||
"PUNC": {POS: PUNCT},
|
||||
"_SP": {POS: SPACE},
|
||||
}
|
||||
|
|
@ -1,7 +1,6 @@
|
|||
from .stop_words import STOP_WORDS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .punctuation import TOKENIZER_SUFFIXES
|
||||
from .tag_map import TAG_MAP
|
||||
|
||||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
from ...language import Language
|
||||
|
|
@ -14,7 +13,6 @@ class UrduDefaults(Language.Defaults):
|
|||
lex_attr_getters[LANG] = lambda text: "ur"
|
||||
|
||||
tokenizer_exceptions = BASE_EXCEPTIONS
|
||||
tag_map = TAG_MAP
|
||||
stop_words = STOP_WORDS
|
||||
suffixes = TOKENIZER_SUFFIXES
|
||||
writing_system = {"direction": "rtl", "has_case": False, "has_letters": True}
|
||||
|
|
|
|||
|
|
@ -1,95 +0,0 @@
|
|||
from ...symbols import NOUN, PROPN, PART, INTJ, SPACE, PRON, AUX, SCONJ
|
||||
from ...symbols import POS, PUNCT, SYM, ADJ, CCONJ, NUM, DET, ADV, ADP, X, VERB
|
||||
|
||||
TAG_MAP = {
|
||||
"JJ-Ez": {POS: ADJ},
|
||||
"INJC": {POS: X},
|
||||
"QFC": {POS: DET},
|
||||
"UNK": {POS: X},
|
||||
"NSTC": {POS: ADV},
|
||||
"NST": {POS: ADV},
|
||||
"VMC": {POS: VERB},
|
||||
"PRPC": {POS: PRON},
|
||||
"RBC": {POS: ADV},
|
||||
"PSPC": {POS: ADP},
|
||||
"INJ": {POS: X},
|
||||
"JJZ": {POS: ADJ},
|
||||
"CCC": {POS: SCONJ},
|
||||
"NN-Ez": {POS: NOUN},
|
||||
"ECH": {POS: NOUN},
|
||||
"WQ": {POS: DET},
|
||||
"RDP": {POS: ADJ},
|
||||
"JJC": {POS: ADJ},
|
||||
"NEG": {POS: PART},
|
||||
"NNZ": {POS: NOUN},
|
||||
"QO": {POS: ADJ},
|
||||
"INTFC": {POS: ADV},
|
||||
"INTF": {POS: ADV},
|
||||
"NFC": {POS: ADP},
|
||||
"QCC": {POS: NUM},
|
||||
"QC": {POS: NUM},
|
||||
"QF": {POS: DET},
|
||||
"VAUX": {POS: AUX},
|
||||
"VM": {POS: VERB},
|
||||
"DEM": {POS: DET},
|
||||
"NNPC": {POS: PROPN},
|
||||
"NNC": {POS: NOUN},
|
||||
"PSP": {POS: ADP},
|
||||
".": {POS: PUNCT},
|
||||
",": {POS: PUNCT},
|
||||
"-LRB-": {POS: PUNCT},
|
||||
"-RRB-": {POS: PUNCT},
|
||||
"``": {POS: PUNCT},
|
||||
'""': {POS: PUNCT},
|
||||
"''": {POS: PUNCT},
|
||||
":": {POS: PUNCT},
|
||||
"$": {POS: SYM},
|
||||
"#": {POS: SYM},
|
||||
"AFX": {POS: ADJ},
|
||||
"CC": {POS: CCONJ},
|
||||
"CD": {POS: NUM},
|
||||
"DT": {POS: DET},
|
||||
"EX": {POS: ADV},
|
||||
"FW": {POS: X},
|
||||
"HYPH": {POS: PUNCT},
|
||||
"IN": {POS: ADP},
|
||||
"JJ": {POS: ADJ},
|
||||
"JJR": {POS: ADJ},
|
||||
"JJS": {POS: ADJ},
|
||||
"LS": {POS: PUNCT},
|
||||
"MD": {POS: VERB},
|
||||
"NIL": {POS: ""},
|
||||
"NN": {POS: NOUN},
|
||||
"NNP": {POS: PROPN},
|
||||
"NNPS": {POS: PROPN},
|
||||
"NNS": {POS: NOUN},
|
||||
"PDT": {POS: ADJ},
|
||||
"POS": {POS: PART},
|
||||
"PRP": {POS: PRON},
|
||||
"PRP$": {POS: ADJ},
|
||||
"RB": {POS: ADV},
|
||||
"RBR": {POS: ADV},
|
||||
"RBS": {POS: ADV},
|
||||
"RP": {POS: PART},
|
||||
"SP": {POS: SPACE},
|
||||
"SYM": {POS: SYM},
|
||||
"TO": {POS: PART},
|
||||
"UH": {POS: INTJ},
|
||||
"VB": {POS: VERB},
|
||||
"VBD": {POS: VERB},
|
||||
"VBG": {POS: VERB},
|
||||
"VBN": {POS: VERB},
|
||||
"VBP": {POS: VERB},
|
||||
"VBZ": {POS: VERB},
|
||||
"WDT": {POS: ADJ},
|
||||
"WP": {POS: NOUN},
|
||||
"WP$": {POS: ADJ},
|
||||
"WRB": {POS: ADV},
|
||||
"ADD": {POS: X},
|
||||
"NFP": {POS: PUNCT},
|
||||
"GW": {POS: X},
|
||||
"XX": {POS: X},
|
||||
"BES": {POS: VERB},
|
||||
"HVS": {POS: VERB},
|
||||
"_SP": {POS: SPACE},
|
||||
}
|
||||
|
|
@ -11,7 +11,6 @@ from ...util import DummyTokenizer
|
|||
from ..tokenizer_exceptions import BASE_EXCEPTIONS
|
||||
from .lex_attrs import LEX_ATTRS
|
||||
from .stop_words import STOP_WORDS
|
||||
from .tag_map import TAG_MAP
|
||||
from ... import util
|
||||
|
||||
|
||||
|
|
@ -300,7 +299,6 @@ class ChineseDefaults(Language.Defaults):
|
|||
lex_attr_getters[LANG] = lambda text: "zh"
|
||||
tokenizer_exceptions = BASE_EXCEPTIONS
|
||||
stop_words = STOP_WORDS
|
||||
tag_map = TAG_MAP
|
||||
writing_system = {"direction": "ltr", "has_case": False, "has_letters": False}
|
||||
|
||||
@classmethod
|
||||
|
|
|
|||
|
|
@ -1,46 +0,0 @@
|
|||
from ...symbols import POS, PUNCT, ADJ, SCONJ, CCONJ, NUM, DET, ADV, ADP, X
|
||||
from ...symbols import NOUN, PART, INTJ, PRON, VERB, SPACE, PROPN
|
||||
|
||||
# The Chinese part-of-speech tagger uses the OntoNotes 5 version of the Penn
|
||||
# Treebank tag set. We also map the tags to the simpler Universal Dependencies
|
||||
# v2 tag set.
|
||||
|
||||
TAG_MAP = {
|
||||
"AS": {POS: PART},
|
||||
"DEC": {POS: PART},
|
||||
"DEG": {POS: PART},
|
||||
"DER": {POS: PART},
|
||||
"DEV": {POS: PART},
|
||||
"ETC": {POS: PART},
|
||||
"LC": {POS: PART},
|
||||
"MSP": {POS: PART},
|
||||
"SP": {POS: PART},
|
||||
"BA": {POS: X},
|
||||
"FW": {POS: X},
|
||||
"IJ": {POS: INTJ},
|
||||
"LB": {POS: X},
|
||||
"ON": {POS: X},
|
||||
"SB": {POS: X},
|
||||
"X": {POS: X},
|
||||
"URL": {POS: X},
|
||||
"INF": {POS: X},
|
||||
"NN": {POS: NOUN},
|
||||
"NR": {POS: PROPN},
|
||||
"NT": {POS: NOUN},
|
||||
"VA": {POS: VERB},
|
||||
"VC": {POS: VERB},
|
||||
"VE": {POS: VERB},
|
||||
"VV": {POS: VERB},
|
||||
"CD": {POS: NUM},
|
||||
"M": {POS: NUM},
|
||||
"OD": {POS: NUM},
|
||||
"DT": {POS: DET},
|
||||
"CC": {POS: CCONJ},
|
||||
"CS": {POS: SCONJ},
|
||||
"AD": {POS: ADV},
|
||||
"JJ": {POS: ADJ},
|
||||
"P": {POS: ADP},
|
||||
"PN": {POS: PRON},
|
||||
"PU": {POS: PUNCT},
|
||||
"_SP": {POS: SPACE},
|
||||
}
|
||||
|
|
@ -78,9 +78,7 @@ class BaseDefaults:
|
|||
BASE_NORMS,
|
||||
vocab.lookups.get_table("lexeme_norm"),
|
||||
)
|
||||
for tag_str, exc in cls.morph_rules.items():
|
||||
for orth_str, attrs in exc.items():
|
||||
vocab.morphology.add_special_case(tag_str, orth_str, attrs)
|
||||
vocab.morphology.load_morph_exceptions(cls.morph_rules)
|
||||
return vocab
|
||||
|
||||
@classmethod
|
||||
|
|
|
|||
|
|
@ -20,7 +20,7 @@ cdef class Morphology:
|
|||
cdef readonly object tag_map
|
||||
cdef readonly object tag_names
|
||||
cdef readonly object reverse_index
|
||||
cdef readonly object exc
|
||||
cdef readonly object _exc
|
||||
cdef readonly PreshMapArray _cache
|
||||
cdef readonly int n_tags
|
||||
|
||||
|
|
|
|||
|
|
@ -68,12 +68,9 @@ cdef class Morphology:
|
|||
self.lemmatizer = lemmatizer
|
||||
|
||||
self._cache = PreshMapArray(self.n_tags)
|
||||
self.exc = {}
|
||||
self._exc = {}
|
||||
if exc is not None:
|
||||
for (tag, orth), attrs in exc.items():
|
||||
attrs = _normalize_props(attrs)
|
||||
self.add_special_case(
|
||||
self.strings.as_string(tag), self.strings.as_string(orth), attrs)
|
||||
self.load_morph_exceptions(exc)
|
||||
|
||||
def load_tag_map(self, tag_map):
|
||||
self.tag_map = {}
|
||||
|
|
@ -99,7 +96,7 @@ cdef class Morphology:
|
|||
|
||||
def __reduce__(self):
|
||||
return (Morphology, (self.strings, self.tag_map, self.lemmatizer,
|
||||
self.exc), None, None)
|
||||
self._exc), None, None)
|
||||
|
||||
def add(self, features):
|
||||
"""Insert a morphological analysis in the morphology table, if not
|
||||
|
|
@ -224,7 +221,7 @@ cdef class Morphology:
|
|||
attrs = _normalize_props(attrs)
|
||||
self.add(attrs)
|
||||
attrs = intify_attrs(attrs, self.strings, _do_deprecated=True)
|
||||
self.exc[(tag_str, self.strings.add(orth_str))] = attrs
|
||||
self._exc[(tag_str, self.strings.add(orth_str))] = attrs
|
||||
|
||||
cdef int assign_untagged(self, TokenC* token) except -1:
|
||||
"""Set morphological attributes on a token without a POS tag. Uses
|
||||
|
|
@ -270,21 +267,34 @@ cdef class Morphology:
|
|||
token.pos = <univ_pos_t>pos
|
||||
token.tag = self.strings[tag_str]
|
||||
token.morph = self.add(features)
|
||||
if (self.tag_names[tag_id], token.lex.orth) in self.exc:
|
||||
if (self.tag_names[tag_id], token.lex.orth) in self._exc:
|
||||
self._assign_tag_from_exceptions(token, tag_id)
|
||||
|
||||
cdef int _assign_tag_from_exceptions(self, TokenC* token, int tag_id) except -1:
|
||||
key = (self.tag_names[tag_id], token.lex.orth)
|
||||
cdef dict attrs
|
||||
attrs = self.exc[key]
|
||||
attrs = self._exc[key]
|
||||
token.pos = attrs.get(POS, token.pos)
|
||||
token.lemma = attrs.get(LEMMA, token.lemma)
|
||||
|
||||
def load_morph_exceptions(self, dict exc):
|
||||
def load_morph_exceptions(self, dict morph_rules):
|
||||
self._exc = {}
|
||||
# Map (form, pos) to attributes
|
||||
for tag_str, entries in exc.items():
|
||||
for form_str, attrs in entries.items():
|
||||
self.add_special_case(tag_str, form_str, attrs)
|
||||
for tag, exc in morph_rules.items():
|
||||
for orth, attrs in exc.items():
|
||||
attrs = _normalize_props(attrs)
|
||||
self.add_special_case(self.strings.as_string(tag), self.strings.as_string(orth), attrs)
|
||||
|
||||
@property
|
||||
def exc(self):
|
||||
# generate the serializable exc in the MORPH_RULES format from the
|
||||
# internal tuple-key format
|
||||
morph_rules = {}
|
||||
for (tag, orth) in sorted(self._exc):
|
||||
if not tag in morph_rules:
|
||||
morph_rules[tag] = {}
|
||||
morph_rules[tag][self.strings[orth]] = self._exc[(tag, orth)]
|
||||
return morph_rules
|
||||
|
||||
@staticmethod
|
||||
def feats_to_dict(feats):
|
||||
|
|
|
|||
|
|
@ -359,9 +359,7 @@ class Tagger(Pipe):
|
|||
if new_tag_map:
|
||||
if "_SP" in orig_tag_map:
|
||||
new_tag_map["_SP"] = orig_tag_map["_SP"]
|
||||
vocab.morphology = Morphology(vocab.strings, new_tag_map,
|
||||
vocab.morphology.lemmatizer,
|
||||
exc=vocab.morphology.exc)
|
||||
vocab.morphology.load_tag_map(new_tag_map)
|
||||
self.set_output(len(self.labels))
|
||||
doc_sample = [Doc(self.vocab, words=["hello", "world"])]
|
||||
if pipeline is not None:
|
||||
|
|
@ -400,10 +398,7 @@ class Tagger(Pipe):
|
|||
if values is None:
|
||||
values = {POS: "X"}
|
||||
tag_map[label] = values
|
||||
self.vocab.morphology = Morphology(
|
||||
self.vocab.strings, tag_map=tag_map,
|
||||
lemmatizer=self.vocab.morphology.lemmatizer,
|
||||
exc=self.vocab.morphology.exc)
|
||||
self.vocab.morphology.load_tag_map(tag_map)
|
||||
return 1
|
||||
|
||||
def use_params(self, params):
|
||||
|
|
@ -417,6 +412,8 @@ class Tagger(Pipe):
|
|||
serialize["cfg"] = lambda: srsly.json_dumps(self.cfg)
|
||||
tag_map = dict(sorted(self.vocab.morphology.tag_map.items()))
|
||||
serialize["tag_map"] = lambda: srsly.msgpack_dumps(tag_map)
|
||||
morph_rules = dict(self.vocab.morphology.exc)
|
||||
serialize["morph_rules"] = lambda: srsly.msgpack_dumps(morph_rules)
|
||||
return util.to_bytes(serialize, exclude)
|
||||
|
||||
def from_bytes(self, bytes_data, exclude=tuple()):
|
||||
|
|
@ -428,14 +425,18 @@ class Tagger(Pipe):
|
|||
|
||||
def load_tag_map(b):
|
||||
tag_map = srsly.msgpack_loads(b)
|
||||
self.vocab.morphology = Morphology(
|
||||
self.vocab.strings, tag_map=tag_map,
|
||||
lemmatizer=self.vocab.morphology.lemmatizer,
|
||||
exc=self.vocab.morphology.exc)
|
||||
self.vocab.morphology.load_tag_map(tag_map)
|
||||
|
||||
def load_morph_rules(b):
|
||||
morph_rules = srsly.msgpack_loads(b)
|
||||
self.vocab.morphology.load_morph_exceptions(morph_rules)
|
||||
|
||||
self.vocab.morphology = Morphology(self.vocab.strings, dict(),
|
||||
lemmatizer=self.vocab.morphology.lemmatizer)
|
||||
deserialize = {
|
||||
"vocab": lambda b: self.vocab.from_bytes(b),
|
||||
"tag_map": load_tag_map,
|
||||
"morph_rules": load_morph_rules,
|
||||
"cfg": lambda b: self.cfg.update(srsly.json_loads(b)),
|
||||
"model": lambda b: load_model(b),
|
||||
}
|
||||
|
|
@ -444,9 +445,11 @@ class Tagger(Pipe):
|
|||
|
||||
def to_disk(self, path, exclude=tuple()):
|
||||
tag_map = dict(sorted(self.vocab.morphology.tag_map.items()))
|
||||
morph_rules = dict(self.vocab.morphology.exc)
|
||||
serialize = {
|
||||
"vocab": lambda p: self.vocab.to_disk(p),
|
||||
"tag_map": lambda p: srsly.write_msgpack(p, tag_map),
|
||||
"morph_rules": lambda p: srsly.write_msgpack(p, morph_rules),
|
||||
"model": lambda p: self.model.to_disk(p),
|
||||
"cfg": lambda p: srsly.write_json(p, self.cfg),
|
||||
}
|
||||
|
|
@ -462,15 +465,19 @@ class Tagger(Pipe):
|
|||
|
||||
def load_tag_map(p):
|
||||
tag_map = srsly.read_msgpack(p)
|
||||
self.vocab.morphology = Morphology(
|
||||
self.vocab.strings, tag_map=tag_map,
|
||||
lemmatizer=self.vocab.morphology.lemmatizer,
|
||||
exc=self.vocab.morphology.exc)
|
||||
self.vocab.morphology.load_tag_map(tag_map)
|
||||
|
||||
def load_morph_rules(p):
|
||||
morph_rules = srsly.read_msgpack(p)
|
||||
self.vocab.morphology.load_morph_exceptions(morph_rules)
|
||||
|
||||
self.vocab.morphology = Morphology(self.vocab.strings, dict(),
|
||||
lemmatizer=self.vocab.morphology.lemmatizer)
|
||||
deserialize = {
|
||||
"vocab": lambda p: self.vocab.from_disk(p),
|
||||
"cfg": lambda p: self.cfg.update(_load_cfg(p)),
|
||||
"tag_map": load_tag_map,
|
||||
"morph_rules": load_morph_rules,
|
||||
"model": load_model,
|
||||
}
|
||||
util.from_disk(path, deserialize, exclude)
|
||||
|
|
|
|||
|
|
@ -1,9 +1,21 @@
|
|||
import pytest
|
||||
from spacy.symbols import POS, PRON, VERB
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def i_has(en_tokenizer):
|
||||
doc = en_tokenizer("I has")
|
||||
tag_map = {
|
||||
"PRP": {POS: PRON, "PronType": "prs"},
|
||||
"VBZ": {
|
||||
POS: VERB,
|
||||
"VerbForm": "fin",
|
||||
"Tense": "pres",
|
||||
"Number": "sing",
|
||||
"Person": "three",
|
||||
},
|
||||
}
|
||||
en_tokenizer.vocab.morphology.load_tag_map(tag_map)
|
||||
doc[0].tag_ = "PRP"
|
||||
doc[1].tag_ = "VBZ"
|
||||
return doc
|
||||
|
|
|
|||
|
|
@ -4,11 +4,11 @@ from ...util import get_doc
|
|||
def test_de_parser_noun_chunks_standard_de(de_tokenizer):
|
||||
text = "Eine Tasse steht auf dem Tisch."
|
||||
heads = [1, 1, 0, -1, 1, -2, -4]
|
||||
tags = ["ART", "NN", "VVFIN", "APPR", "ART", "NN", "$."]
|
||||
pos = ["DET", "NOUN", "VERB", "ADP", "DET", "NOUN", "PUNCT"]
|
||||
deps = ["nk", "sb", "ROOT", "mo", "nk", "nk", "punct"]
|
||||
tokens = de_tokenizer(text)
|
||||
doc = get_doc(
|
||||
tokens.vocab, words=[t.text for t in tokens], tags=tags, deps=deps, heads=heads
|
||||
tokens.vocab, words=[t.text for t in tokens], pos=pos, deps=deps, heads=heads
|
||||
)
|
||||
chunks = list(doc.noun_chunks)
|
||||
assert len(chunks) == 2
|
||||
|
|
@ -19,11 +19,11 @@ def test_de_parser_noun_chunks_standard_de(de_tokenizer):
|
|||
def test_de_extended_chunk(de_tokenizer):
|
||||
text = "Die Sängerin singt mit einer Tasse Kaffee Arien."
|
||||
heads = [1, 1, 0, -1, 1, -2, -1, -5, -6]
|
||||
tags = ["ART", "NN", "VVFIN", "APPR", "ART", "NN", "NN", "NN", "$."]
|
||||
pos = ["DET", "NOUN", "VERB", "ADP", "DET", "NOUN", "NOUN", "NOUN", "PUNCT"]
|
||||
deps = ["nk", "sb", "ROOT", "mo", "nk", "nk", "nk", "oa", "punct"]
|
||||
tokens = de_tokenizer(text)
|
||||
doc = get_doc(
|
||||
tokens.vocab, words=[t.text for t in tokens], tags=tags, deps=deps, heads=heads
|
||||
tokens.vocab, words=[t.text for t in tokens], pos=pos, deps=deps, heads=heads
|
||||
)
|
||||
chunks = list(doc.noun_chunks)
|
||||
assert len(chunks) == 3
|
||||
|
|
|
|||
|
|
@ -4,11 +4,11 @@ from ...util import get_doc
|
|||
def test_en_parser_noun_chunks_standard(en_tokenizer):
|
||||
text = "A base phrase should be recognized."
|
||||
heads = [2, 1, 3, 2, 1, 0, -1]
|
||||
tags = ["DT", "JJ", "NN", "MD", "VB", "VBN", "."]
|
||||
pos = ["DET", "ADJ", "NOUN", "AUX", "VERB", "VERB", "PUNCT"]
|
||||
deps = ["det", "amod", "nsubjpass", "aux", "auxpass", "ROOT", "punct"]
|
||||
tokens = en_tokenizer(text)
|
||||
doc = get_doc(
|
||||
tokens.vocab, words=[t.text for t in tokens], tags=tags, deps=deps, heads=heads
|
||||
tokens.vocab, words=[t.text for t in tokens], pos=pos, deps=deps, heads=heads
|
||||
)
|
||||
chunks = list(doc.noun_chunks)
|
||||
assert len(chunks) == 1
|
||||
|
|
@ -19,12 +19,12 @@ def test_en_parser_noun_chunks_coordinated(en_tokenizer):
|
|||
# fmt: off
|
||||
text = "A base phrase and a good phrase are often the same."
|
||||
heads = [2, 1, 5, -1, 2, 1, -4, 0, -1, 1, -3, -4]
|
||||
tags = ["DT", "NN", "NN", "CC", "DT", "JJ", "NN", "VBP", "RB", "DT", "JJ", "."]
|
||||
pos = ["DET", "NOUN", "NOUN", "CCONJ", "DET", "ADJ", "NOUN", "VERB", "ADV", "DET", "ADJ", "PUNCT"]
|
||||
deps = ["det", "compound", "nsubj", "cc", "det", "amod", "conj", "ROOT", "advmod", "det", "attr", "punct"]
|
||||
# fmt: on
|
||||
tokens = en_tokenizer(text)
|
||||
doc = get_doc(
|
||||
tokens.vocab, words=[t.text for t in tokens], tags=tags, deps=deps, heads=heads
|
||||
tokens.vocab, words=[t.text for t in tokens], pos=pos, deps=deps, heads=heads
|
||||
)
|
||||
chunks = list(doc.noun_chunks)
|
||||
assert len(chunks) == 2
|
||||
|
|
@ -35,11 +35,11 @@ def test_en_parser_noun_chunks_coordinated(en_tokenizer):
|
|||
def test_en_parser_noun_chunks_pp_chunks(en_tokenizer):
|
||||
text = "A phrase with another phrase occurs."
|
||||
heads = [1, 4, -1, 1, -2, 0, -1]
|
||||
tags = ["DT", "NN", "IN", "DT", "NN", "VBZ", "."]
|
||||
pos = ["DET", "NOUN", "ADP", "DET", "NOUN", "VERB", "PUNCT"]
|
||||
deps = ["det", "nsubj", "prep", "det", "pobj", "ROOT", "punct"]
|
||||
tokens = en_tokenizer(text)
|
||||
doc = get_doc(
|
||||
tokens.vocab, words=[t.text for t in tokens], tags=tags, deps=deps, heads=heads
|
||||
tokens.vocab, words=[t.text for t in tokens], pos=pos, deps=deps, heads=heads
|
||||
)
|
||||
chunks = list(doc.noun_chunks)
|
||||
assert len(chunks) == 2
|
||||
|
|
@ -51,12 +51,12 @@ def test_en_parser_noun_chunks_appositional_modifiers(en_tokenizer):
|
|||
# fmt: off
|
||||
text = "Sam, my brother, arrived to the house."
|
||||
heads = [5, -1, 1, -3, -4, 0, -1, 1, -2, -4]
|
||||
tags = ["NNP", ",", "PRP$", "NN", ",", "VBD", "IN", "DT", "NN", "."]
|
||||
pos = ["PROPN", "PUNCT", "DET", "NOUN", "PUNCT", "VERB", "ADP", "DET", "NOUN", "PUNCT"]
|
||||
deps = ["nsubj", "punct", "poss", "appos", "punct", "ROOT", "prep", "det", "pobj", "punct"]
|
||||
# fmt: on
|
||||
tokens = en_tokenizer(text)
|
||||
doc = get_doc(
|
||||
tokens.vocab, words=[t.text for t in tokens], tags=tags, deps=deps, heads=heads
|
||||
tokens.vocab, words=[t.text for t in tokens], pos=pos, deps=deps, heads=heads
|
||||
)
|
||||
chunks = list(doc.noun_chunks)
|
||||
assert len(chunks) == 3
|
||||
|
|
@ -68,11 +68,11 @@ def test_en_parser_noun_chunks_appositional_modifiers(en_tokenizer):
|
|||
def test_en_parser_noun_chunks_dative(en_tokenizer):
|
||||
text = "She gave Bob a raise."
|
||||
heads = [1, 0, -1, 1, -3, -4]
|
||||
tags = ["PRP", "VBD", "NNP", "DT", "NN", "."]
|
||||
pos = ["PRON", "VERB", "PROPN", "DET", "NOUN", "PUNCT"]
|
||||
deps = ["nsubj", "ROOT", "dative", "det", "dobj", "punct"]
|
||||
tokens = en_tokenizer(text)
|
||||
doc = get_doc(
|
||||
tokens.vocab, words=[t.text for t in tokens], tags=tags, deps=deps, heads=heads
|
||||
tokens.vocab, words=[t.text for t in tokens], pos=pos, deps=deps, heads=heads
|
||||
)
|
||||
chunks = list(doc.noun_chunks)
|
||||
assert len(chunks) == 3
|
||||
|
|
|
|||
|
|
@ -1,10 +1,19 @@
|
|||
from spacy.symbols import POS, PRON, VERB, DET, NOUN, PUNCT
|
||||
from ...util import get_doc
|
||||
|
||||
|
||||
def test_en_tagger_load_morph_exc(en_tokenizer):
|
||||
text = "I like his style."
|
||||
tags = ["PRP", "VBP", "PRP$", "NN", "."]
|
||||
tag_map = {
|
||||
"PRP": {POS: PRON},
|
||||
"VBP": {POS: VERB},
|
||||
"PRP$": {POS: DET},
|
||||
"NN": {POS: NOUN},
|
||||
".": {POS: PUNCT},
|
||||
}
|
||||
morph_exc = {"VBP": {"like": {"lemma": "luck"}}}
|
||||
en_tokenizer.vocab.morphology.load_tag_map(tag_map)
|
||||
en_tokenizer.vocab.morphology.load_morph_exceptions(morph_exc)
|
||||
tokens = en_tokenizer(text)
|
||||
doc = get_doc(tokens.vocab, words=[t.text for t in tokens], tags=tags)
|
||||
|
|
|
|||
|
|
@ -17,6 +17,8 @@ def test_label_types():
|
|||
|
||||
TAG_MAP = {"N": {"pos": "NOUN"}, "V": {"pos": "VERB"}, "J": {"pos": "ADJ"}}
|
||||
|
||||
MORPH_RULES = {"V": {"like": {"lemma": "luck"}}}
|
||||
|
||||
TRAIN_DATA = [
|
||||
("I like green eggs", {"tags": ["N", "V", "J", "N"]}),
|
||||
("Eat blue ham", {"tags": ["V", "J", "N"]}),
|
||||
|
|
@ -26,6 +28,8 @@ TRAIN_DATA = [
|
|||
def test_overfitting_IO():
|
||||
# Simple test to try and quickly overfit the tagger - ensuring the ML models work correctly
|
||||
nlp = English()
|
||||
nlp.vocab.morphology.load_tag_map(TAG_MAP)
|
||||
nlp.vocab.morphology.load_morph_exceptions(MORPH_RULES)
|
||||
tagger = nlp.create_pipe("tagger")
|
||||
nlp.vocab.morphology.load_tag_map(TAG_MAP)
|
||||
train_examples = []
|
||||
|
|
@ -46,6 +50,7 @@ def test_overfitting_IO():
|
|||
assert doc[1].tag_ is "V"
|
||||
assert doc[2].tag_ is "J"
|
||||
assert doc[3].tag_ is "N"
|
||||
assert doc[1].lemma_ == "luck"
|
||||
|
||||
# Also test the results are still the same after IO
|
||||
with make_tempdir() as tmp_dir:
|
||||
|
|
@ -56,3 +61,4 @@ def test_overfitting_IO():
|
|||
assert doc2[1].tag_ is "V"
|
||||
assert doc2[2].tag_ is "J"
|
||||
assert doc2[3].tag_ is "N"
|
||||
assert doc[1].lemma_ == "luck"
|
||||
|
|
|
|||
|
|
@ -27,16 +27,16 @@ def test_issue3002():
|
|||
def test_issue3009(en_vocab):
|
||||
"""Test problem with matcher quantifiers"""
|
||||
patterns = [
|
||||
[{"LEMMA": "have"}, {"LOWER": "to"}, {"LOWER": "do"}, {"TAG": "IN"}],
|
||||
[{"ORTH": "has"}, {"LOWER": "to"}, {"LOWER": "do"}, {"TAG": "IN"}],
|
||||
[
|
||||
{"LEMMA": "have"},
|
||||
{"ORTH": "has"},
|
||||
{"IS_ASCII": True, "IS_PUNCT": False, "OP": "*"},
|
||||
{"LOWER": "to"},
|
||||
{"LOWER": "do"},
|
||||
{"TAG": "IN"},
|
||||
],
|
||||
[
|
||||
{"LEMMA": "have"},
|
||||
{"ORTH": "has"},
|
||||
{"IS_ASCII": True, "IS_PUNCT": False, "OP": "?"},
|
||||
{"LOWER": "to"},
|
||||
{"LOWER": "do"},
|
||||
|
|
@ -45,7 +45,8 @@ def test_issue3009(en_vocab):
|
|||
]
|
||||
words = ["also", "has", "to", "do", "with"]
|
||||
tags = ["RB", "VBZ", "TO", "VB", "IN"]
|
||||
doc = get_doc(en_vocab, words=words, tags=tags)
|
||||
pos = ["ADV", "VERB", "ADP", "VERB", "ADP"]
|
||||
doc = get_doc(en_vocab, words=words, tags=tags, pos=pos)
|
||||
matcher = Matcher(en_vocab)
|
||||
for i, pattern in enumerate(patterns):
|
||||
matcher.add(str(i), [pattern])
|
||||
|
|
|
|||
Loading…
Reference in New Issue
Block a user