Hugging Face's logo

ljvmiranda921
/
xx_cop_sigtyp_trf

Token Classification
spaCy
multilingual
Eval Results
Model card Files Community
ljvmiranda921 commited on
Commit
3b99e24
•
1 Parent(s): c209379

xx_cop_sigtyp_trf-0.1.0: Update SIGTYP pipeline

Browse files
Files changed (21) hide show
  1. .gitattributes +2 -0
  2. README.md +45 -0
  3. config.cfg +190 -0
  4. meta.json +416 -0
  5. parser/cfg +13 -0
  6. parser/model +0 -0
  7. parser/moves +1 -0
  8. tagger/cfg +173 -0
  9. tagger/model +0 -0
  10. tokenizer +3 -0
  11. trainable_lemmatizer/cfg +215 -0
  12. trainable_lemmatizer/model +0 -0
  13. trainable_lemmatizer/trees +0 -0
  14. transformer/cfg +3 -0
  15. transformer/model +3 -0
  16. vocab/key2row +1 -0
  17. vocab/lookups.bin +3 -0
  18. vocab/strings.json +0 -0
  19. vocab/vectors +0 -0
  20. vocab/vectors.cfg +3 -0
  21. xx_cop_sigtyp_trf-any-py3-none-any.whl +3 -0
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ transformer/model filter=lfs diff=lfs merge=lfs -text
37
+ xx_cop_sigtyp_trf-any-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - spacy
4
+ - token-classification
5
+ language:
6
+ - multilingual
7
+ model-index:
8
+ - name: xx_cop_sigtyp_trf
9
+ results:
10
+ - task:
11
+ name: TAG
12
+ type: token-classification
13
+ metrics:
14
+ - name: TAG (XPOS) Accuracy
15
+ type: accuracy
16
+ value: 0.0
17
+ - task:
18
+ name: LEMMA
19
+ type: token-classification
20
+ metrics:
21
+ - name: Lemma Accuracy
22
+ type: accuracy
23
+ value: 0.2837307576
24
+ - task:
25
+ name: UNLABELED_DEPENDENCIES
26
+ type: token-classification
27
+ metrics:
28
+ - name: Unlabeled Attachment Score (UAS)
29
+ type: f_score
30
+ value: 0.0100692259
31
+ - task:
32
+ name: LABELED_DEPENDENCIES
33
+ type: token-classification
34
+ metrics:
35
+ - name: Labeled Attachment Score (LAS)
36
+ type: f_score
37
+ value: 0.0090623033
38
+ - task:
39
+ name: SENTS
40
+ type: token-classification
41
+ metrics:
42
+ - name: Sentences F-Score
43
+ type: f_score
44
+ value: 0.0109809663
45
+ ---
config.cfg ADDED
@@ -0,0 +1,190 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [paths]
2
+ train = "corpus/train/cop_train.spacy"
3
+ dev = "corpus/dev/cop_valid.spacy"
4
+ vectors = null
5
+ init_tok2vec = null
6
+
7
+ [system]
8
+ gpu_allocator = "pytorch"
9
+ seed = 42
10
+
11
+ [nlp]
12
+ lang = "xx"
13
+ pipeline = ["transformer","parser","trainable_lemmatizer","tagger"]
14
+ batch_size = 128
15
+ disabled = []
16
+ before_creation = null
17
+ after_creation = null
18
+ after_pipeline_creation = null
19
+ tokenizer = {"@tokenizers":"spacy.Tokenizer.v1"}
20
+
21
+ [components]
22
+
23
+ [components.parser]
24
+ factory = "parser"
25
+ learn_tokens = false
26
+ min_action_freq = 30
27
+ moves = null
28
+ scorer = {"@scorers":"spacy.parser_scorer.v1"}
29
+ update_with_oracle_cut_size = 100
30
+
31
+ [components.parser.model]
32
+ @architectures = "spacy.TransitionBasedParser.v2"
33
+ state_type = "parser"
34
+ extra_state_tokens = false
35
+ hidden_width = 128
36
+ maxout_pieces = 3
37
+ use_upper = false
38
+ nO = null
39
+
40
+ [components.parser.model.tok2vec]
41
+ @architectures = "spacy-transformers.TransformerListener.v1"
42
+ grad_factor = 1.0
43
+ pooling = {"@layers":"reduce_mean.v1"}
44
+ upstream = "*"
45
+
46
+ [components.tagger]
47
+ factory = "tagger"
48
+ label_smoothing = 0.0
49
+ neg_prefix = "!"
50
+ overwrite = false
51
+ scorer = {"@scorers":"spacy.tagger_scorer.v1"}
52
+
53
+ [components.tagger.model]
54
+ @architectures = "spacy.Tagger.v2"
55
+ nO = null
56
+ normalize = false
57
+
58
+ [components.tagger.model.tok2vec]
59
+ @architectures = "spacy-transformers.TransformerListener.v1"
60
+ grad_factor = 1.0
61
+ pooling = {"@layers":"reduce_mean.v1"}
62
+ upstream = "*"
63
+
64
+ [components.trainable_lemmatizer]
65
+ factory = "trainable_lemmatizer"
66
+ backoff = "orth"
67
+ min_tree_freq = 3
68
+ overwrite = false
69
+ scorer = {"@scorers":"spacy.lemmatizer_scorer.v1"}
70
+ top_k = 1
71
+
72
+ [components.trainable_lemmatizer.model]
73
+ @architectures = "spacy.Tagger.v2"
74
+ nO = null
75
+ normalize = false
76
+
77
+ [components.trainable_lemmatizer.model.tok2vec]
78
+ @architectures = "spacy-transformers.TransformerListener.v1"
79
+ grad_factor = 1.0
80
+ pooling = {"@layers":"reduce_mean.v1"}
81
+ upstream = "*"
82
+
83
+ [components.transformer]
84
+ factory = "transformer"
85
+ max_batch_items = 4096
86
+ set_extra_annotations = {"@annotation_setters":"spacy-transformers.null_annotation_setter.v1"}
87
+
88
+ [components.transformer.model]
89
+ @architectures = "spacy-transformers.TransformerModel.v3"
90
+ name = "ljvmiranda921/LiBERTus-base"
91
+ mixed_precision = false
92
+
93
+ [components.transformer.model.get_spans]
94
+ @span_getters = "spacy-transformers.strided_spans.v1"
95
+ window = 128
96
+ stride = 96
97
+
98
+ [components.transformer.model.grad_scaler_config]
99
+
100
+ [components.transformer.model.tokenizer_config]
101
+ use_fast = true
102
+
103
+ [components.transformer.model.transformer_config]
104
+
105
+ [corpora]
106
+
107
+ [corpora.dev]
108
+ @readers = "spacy.Corpus.v1"
109
+ path = ${paths.dev}
110
+ max_length = 0
111
+ gold_preproc = false
112
+ limit = 0
113
+ augmenter = null
114
+
115
+ [corpora.train]
116
+ @readers = "spacy.Corpus.v1"
117
+ path = ${paths.train}
118
+ max_length = 0
119
+ gold_preproc = false
120
+ limit = 0
121
+ augmenter = null
122
+
123
+ [training]
124
+ accumulate_gradient = 3
125
+ dev_corpus = "corpora.dev"
126
+ train_corpus = "corpora.train"
127
+ seed = ${system.seed}
128
+ gpu_allocator = ${system.gpu_allocator}
129
+ dropout = 0.1
130
+ patience = 1600
131
+ max_epochs = 0
132
+ max_steps = 20000
133
+ eval_frequency = 200
134
+ frozen_components = []
135
+ annotating_components = []
136
+ before_to_disk = null
137
+ before_update = null
138
+
139
+ [training.batcher]
140
+ @batchers = "spacy.batch_by_padded.v1"
141
+ discard_oversize = true
142
+ size = 2000
143
+ buffer = 256
144
+ get_length = null
145
+
146
+ [training.logger]
147
+ @loggers = "spacy.ConsoleLogger.v3"
148
+ console_output = true
149
+ progress_bar = "eval"
150
+ output_file = null
151
+
152
+ [training.optimizer]
153
+ @optimizers = "Adam.v1"
154
+ beta1 = 0.9
155
+ beta2 = 0.999
156
+ L2_is_weight_decay = true
157
+ L2 = 0.01
158
+ grad_clip = 1.0
159
+ use_averages = false
160
+ eps = 0.00000001
161
+
162
+ [training.optimizer.learn_rate]
163
+ @schedules = "warmup_linear.v1"
164
+ warmup_steps = 250
165
+ total_steps = 20000
166
+ initial_rate = 0.00005
167
+
168
+ [training.score_weights]
169
+ dep_uas = 0.17
170
+ dep_las = 0.17
171
+ dep_las_per_type = null
172
+ sents_p = null
173
+ sents_r = null
174
+ sents_f = 0.0
175
+ lemma_acc = 0.33
176
+ tag_acc = 0.33
177
+
178
+ [pretraining]
179
+
180
+ [initialize]
181
+ vectors = ${paths.vectors}
182
+ init_tok2vec = ${paths.init_tok2vec}
183
+ vocab_data = null
184
+ lookups = null
185
+ before_init = null
186
+ after_init = null
187
+
188
+ [initialize.components]
189
+
190
+ [initialize.tokenizer]
meta.json ADDED
@@ -0,0 +1,416 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "lang":"xx",
3
+ "name":"cop_sigtyp_trf",
4
+ "version":"0.1.0",
5
+ "description":"",
6
+ "author":"",
7
+ "email":"",
8
+ "url":"",
9
+ "license":"",
10
+ "spacy_version":">=3.6.1,<3.7.0",
11
+ "spacy_git_version":"458bc5f45",
12
+ "vectors":{
13
+ "width":0,
14
+ "vectors":0,
15
+ "keys":0,
16
+ "name":null
17
+ },
18
+ "labels":{
19
+ "transformer":[
20
+
21
+ ],
22
+ "parser":[
23
+ "ROOT",
24
+ "acl:relcl",
25
+ "advcl",
26
+ "advmod",
27
+ "appos",
28
+ "aux",
29
+ "case",
30
+ "cc",
31
+ "ccomp",
32
+ "conj",
33
+ "cop",
34
+ "csubj",
35
+ "dep",
36
+ "det",
37
+ "discourse",
38
+ "dislocated",
39
+ "fixed",
40
+ "flat",
41
+ "iobj",
42
+ "mark",
43
+ "nmod",
44
+ "nsubj",
45
+ "nummod",
46
+ "obj",
47
+ "obl",
48
+ "obl:npmod",
49
+ "parataxis",
50
+ "punct",
51
+ "vocative",
52
+ "xcomp"
53
+ ],
54
+ "tagger":[
55
+ "AAOR",
56
+ "ACAUS",
57
+ "ACOND",
58
+ "ACOND_PPERS",
59
+ "ACOND_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
60
+ "ACOND_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
61
+ "ACOND_PPERS__Definite=Def|Gender=Masc|Number=Sing|Person=2|PronType=Prs",
62
+ "ACOND_PPERS__Definite=Def|Gender=Masc|Number=Sing|Person=3|PronType=Prs",
63
+ "ACOND_PPERS__Definite=Def|Number=Plur|Person=1|PronType=Prs",
64
+ "ACOND_PPERS__Definite=Def|Number=Plur|Person=2|PronType=Prs",
65
+ "ACOND_PPERS__Definite=Def|Number=Plur|Person=3|PronType=Prs",
66
+ "ACOND_PPERS__Definite=Def|Number=Sing|Person=1|PronType=Prs",
67
+ "ACONJ",
68
+ "ACONJ_PPERS",
69
+ "ACONJ_PPERS__Definite=Def|Number=Sing|Person=1|PronType=Prs",
70
+ "ADV",
71
+ "ADV__Foreign=Yes",
72
+ "AFUTCONJ",
73
+ "AJUS",
74
+ "ALIM",
75
+ "ANEGAOR__Polarity=Neg",
76
+ "ANEGJUS__Polarity=Neg",
77
+ "ANEGOPT__Polarity=Neg",
78
+ "ANEGPST_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|Polarity=Neg|PronType=Prs",
79
+ "ANEGPST__Polarity=Neg",
80
+ "ANY__Polarity=Neg",
81
+ "AOPT",
82
+ "AOPT_PPERS",
83
+ "AOPT_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
84
+ "AOPT_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
85
+ "AOPT_PPERS__Definite=Def|Gender=Masc|Number=Sing|Person=2|PronType=Prs",
86
+ "AOPT_PPERS__Definite=Def|Gender=Masc|Number=Sing|Person=3|PronType=Prs",
87
+ "AOPT_PPERS__Definite=Def|Number=Plur|Person=2|PronType=Prs",
88
+ "AOPT_PPERS__Definite=Def|Number=Plur|Person=3|PronType=Prs",
89
+ "AOPT_PPERS__Definite=Def|Number=Sing|Person=1|PronType=Prs",
90
+ "APREC",
91
+ "APST",
92
+ "APST_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
93
+ "ART",
94
+ "ART__Definite=Def|Gender=Fem|Number=Sing|PronType=Art",
95
+ "ART__Definite=Def|Gender=Masc|Number=Sing|PronType=Art",
96
+ "ART__Definite=Def|Number=Plur|PronType=Art",
97
+ "ART__Definite=Ind|Number=Sing|PronType=Art",
98
+ "ART__PronType=Art",
99
+ "CCIRC",
100
+ "CCIRC_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
101
+ "CCIRC_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
102
+ "CFOC",
103
+ "CFOC_PPERS",
104
+ "CFOC_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
105
+ "CONJ",
106
+ "CONJ__Foreign=Yes",
107
+ "CONJ__Foreign=Yes|Polarity=Neg",
108
+ "COP__Gender=Fem|Number=Sing",
109
+ "COP__Gender=Masc|Number=Sing",
110
+ "COP__Number=Plur",
111
+ "CPRET",
112
+ "CPRET_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
113
+ "CREL",
114
+ "CREL_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
115
+ "EXIST",
116
+ "EXIST__Foreign=Yes",
117
+ "EXIST__Polarity=Neg",
118
+ "FM",
119
+ "FM__Foreign=Yes",
120
+ "FM__Foreign=Yes|Polarity=Neg",
121
+ "FM__Polarity=Neg",
122
+ "FUT",
123
+ "IMOD",
124
+ "IMOD_PPERO__Definite=Def|Number=Sing|Person=1|PronType=Prs",
125
+ "IMOD__Definite=Def|PronType=Prs|Reflex=Yes",
126
+ "IMOD__PronType=Tot",
127
+ "N",
128
+ "NEG__Foreign=Yes|Polarity=Neg",
129
+ "NEG__Polarity=Neg",
130
+ "NPROP",
131
+ "NPROP__Foreign=Yes",
132
+ "NUM__Foreign=Yes|NumType=Card",
133
+ "NUM__NumType=Card",
134
+ "N_PPERO__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
135
+ "N_PPERO__Definite=Def|Number=Sing|Person=1|PronType=Prs",
136
+ "N__Foreign=Yes",
137
+ "N__PronType=Rcp",
138
+ "PDEM",
139
+ "PDEM__Definite=Def|Foreign=Yes|Number=Plur|PronType=Dem",
140
+ "PDEM__Definite=Def|Gender=Fem|Number=Sing|PronType=Dem",
141
+ "PDEM__Definite=Def|Gender=Masc|Number=Sing|PronType=Dem",
142
+ "PDEM__Definite=Def|Number=Plur|PronType=Dem",
143
+ "PINT__PronType=Ind",
144
+ "PINT__PronType=Int",
145
+ "PPERI__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
146
+ "PPERI__Definite=Def|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
147
+ "PPERI__Definite=Def|Gender=Masc|Number=Sing|Person=2|PronType=Prs",
148
+ "PPERI__Definite=Def|Gender=Masc|Number=Sing|Person=3|PronType=Prs",
149
+ "PPERI__Definite=Def|Number=Plur|Person=1|PronType=Prs",
150
+ "PPERI__Definite=Def|Number=Plur|Person=2|PronType=Prs",
151
+ "PPERI__Definite=Def|Number=Plur|Person=3|PronType=Prs",
152
+ "PPERI__Definite=Def|Number=Sing|Person=1|PronType=Prs",
153
+ "PPERO",
154
+ "PPERO__Definite=Def|Foreign=Yes|Number=Plur|Person=3|PronType=Prs",
155
+ "PPERO__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
156
+ "PPERO__Definite=Def|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
157
+ "PPERO__Definite=Def|Gender=Masc|Number=Sing|Person=2|PronType=Prs",
158
+ "PPERO__Definite=Def|Gender=Masc|Number=Sing|Person=3|PronType=Prs",
159
+ "PPERO__Definite=Def|Number=Plur|Person=1|PronType=Prs",
160
+ "PPERO__Definite=Def|Number=Plur|Person=2|PronType=Prs",
161
+ "PPERO__Definite=Def|Number=Plur|Person=3|PronType=Prs",
162
+ "PPERO__Definite=Def|Number=Sing|Person=1|PronType=Prs",
163
+ "PPERO__Definite=Def|Poss=Yes|PronType=Prs",
164
+ "PPERS",
165
+ "PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
166
+ "PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
167
+ "PPERS__Definite=Def|Gender=Masc|Number=Sing|Person=2|PronType=Prs",
168
+ "PPERS__Definite=Def|Gender=Masc|Number=Sing|Person=3|PronType=Prs",
169
+ "PPERS__Definite=Def|Number=Plur|Person=1|PronType=Prs",
170
+ "PPERS__Definite=Def|Number=Plur|Person=2|PronType=Prs",
171
+ "PPERS__Definite=Def|Number=Plur|Person=3|PronType=Prs",
172
+ "PPERS__Definite=Def|Number=Sing|Person=1|PronType=Prs",
173
+ "PPOS",
174
+ "PPOS__Definite=Def|Gender=Fem|Gender[psor]=Fem|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs",
175
+ "PPOS__Definite=Def|Gender=Fem|Gender[psor]=Fem|Number=Sing|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs",
176
+ "PPOS__Definite=Def|Gender=Fem|Gender[psor]=Masc|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs",
177
+ "PPOS__Definite=Def|Gender=Fem|Gender[psor]=Masc|Number=Sing|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs",
178
+ "PPOS__Definite=Def|Gender=Fem|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs",
179
+ "PPOS__Definite=Def|Gender=Fem|Number=Sing|Number[psor]=Plur|Person=2|Poss=Yes|PronType=Prs",
180
+ "PPOS__Definite=Def|Gender=Fem|Number=Sing|Number[psor]=Plur|Person=3|Poss=Yes|PronType=Prs",
181
+ "PPOS__Definite=Def|Gender=Fem|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs",
182
+ "PPOS__Definite=Def|Gender=Fem|Number=Sing|Poss=Yes|PronType=Prs",
183
+ "PPOS__Definite=Def|Gender=Masc|Gender[psor]=Fem|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs",
184
+ "PPOS__Definite=Def|Gender=Masc|Gender[psor]=Fem|Number=Sing|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs",
185
+ "PPOS__Definite=Def|Gender=Masc|Gender[psor]=Masc|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs",
186
+ "PPOS__Definite=Def|Gender=Masc|Gender[psor]=Masc|Number=Sing|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs",
187
+ "PPOS__Definite=Def|Gender=Masc|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs",
188
+ "PPOS__Definite=Def|Gender=Masc|Number=Sing|Number[psor]=Plur|Person=2|Poss=Yes|PronType=Prs",
189
+ "PPOS__Definite=Def|Gender=Masc|Number=Sing|Number[psor]=Plur|Person=3|Poss=Yes|PronType=Prs",
190
+ "PPOS__Definite=Def|Gender=Masc|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs",
191
+ "PPOS__Definite=Def|Gender=Masc|Number=Sing|Poss=Yes|PronType=Prs",
192
+ "PPOS__Definite=Def|Gender[psor]=Fem|Number=Plur|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs",
193
+ "PPOS__Definite=Def|Gender[psor]=Fem|Number=Plur|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs",
194
+ "PPOS__Definite=Def|Gender[psor]=Masc|Number=Plur|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs",
195
+ "PPOS__Definite=Def|Gender[psor]=Masc|Number=Plur|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs",
196
+ "PPOS__Definite=Def|Number=Plur|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs",
197
+ "PPOS__Definite=Def|Number=Plur|Number[psor]=Plur|Person=2|Poss=Yes|PronType=Prs",
198
+ "PPOS__Definite=Def|Number=Plur|Number[psor]=Plur|Person=3|Poss=Yes|PronType=Prs",
199
+ "PPOS__Definite=Def|Number=Plur|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs",
200
+ "PPOS__Definite=Def|Number=Plur|Poss=Yes|PronType=Prs",
201
+ "PREP",
202
+ "PREP_PPERO",
203
+ "PREP_PPERO__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
204
+ "PREP_PPERO__Definite=Def|Number=Sing|Person=1|PronType=Prs",
205
+ "PREP__Foreign=Yes",
206
+ "PTC",
207
+ "PTC__Foreign=Yes",
208
+ "PTC__Polarity=Neg",
209
+ "PUNCT",
210
+ "UNKNOWN",
211
+ "UNKNOWN__Foreign=Yes",
212
+ "VBD__VerbForm=Fin",
213
+ "VIMP__Foreign=Yes|VerbForm=Fin",
214
+ "VIMP__VerbForm=Fin",
215
+ "VSTAT__VerbForm=Fin",
216
+ "V_PPERO__VerbForm=Fin",
217
+ "V__Foreign=Yes|VerbForm=Fin",
218
+ "V__Foreign=Yes|VerbForm=Inf",
219
+ "V__VerbForm=Fin",
220
+ "V__VerbForm=Inf"
221
+ ]
222
+ },
223
+ "pipeline":[
224
+ "transformer",
225
+ "parser",
226
+ "trainable_lemmatizer",
227
+ "tagger"
228
+ ],
229
+ "components":[
230
+ "transformer",
231
+ "parser",
232
+ "trainable_lemmatizer",
233
+ "tagger"
234
+ ],
235
+ "disabled":[
236
+
237
+ ],
238
+ "performance":{
239
+ "dep_uas":0.0100692259,
240
+ "dep_las":0.0090623033,
241
+ "dep_las_per_type":{
242
+ "cc":{
243
+ "p":0.0,
244
+ "r":0.0,
245
+ "f":0.0
246
+ },
247
+ "mark":{
248
+ "p":0.0,
249
+ "r":0.0,
250
+ "f":0.0
251
+ },
252
+ "nsubj":{
253
+ "p":0.0,
254
+ "r":0.0,
255
+ "f":0.0
256
+ },
257
+ "root":{
258
+ "p":0.0379346681,
259
+ "r":0.1666666667,
260
+ "f":0.0618025751
261
+ },
262
+ "case":{
263
+ "p":0.0,
264
+ "r":0.0,
265
+ "f":0.0
266
+ },
267
+ "obl":{
268
+ "p":0.0,
269
+ "r":0.0,
270
+ "f":0.0
271
+ },
272
+ "acl:relcl":{
273
+ "p":0.0,
274
+ "r":0.0,
275
+ "f":0.0
276
+ },
277
+ "det":{
278
+ "p":0.0,
279
+ "r":0.0,
280
+ "f":0.0
281
+ },
282
+ "aux":{
283
+ "p":0.0,
284
+ "r":0.0,
285
+ "f":0.0
286
+ },
287
+ "dislocated":{
288
+ "p":0.0,
289
+ "r":0.0,
290
+ "f":0.0
291
+ },
292
+ "advmod":{
293
+ "p":0.0,
294
+ "r":0.0,
295
+ "f":0.0
296
+ },
297
+ "advcl":{
298
+ "p":0.0,
299
+ "r":0.0,
300
+ "f":0.0
301
+ },
302
+ "obj":{
303
+ "p":0.0,
304
+ "r":0.0,
305
+ "f":0.0
306
+ },
307
+ "ccomp":{
308
+ "p":0.0,
309
+ "r":0.0,
310
+ "f":0.0
311
+ },
312
+ "parataxis":{
313
+ "p":0.0,
314
+ "r":0.0,
315
+ "f":0.0
316
+ },
317
+ "nmod":{
318
+ "p":0.0,
319
+ "r":0.0,
320
+ "f":0.0
321
+ },
322
+ "cop":{
323
+ "p":0.0,
324
+ "r":0.0,
325
+ "f":0.0
326
+ },
327
+ "appos":{
328
+ "p":0.0,
329
+ "r":0.0,
330
+ "f":0.0
331
+ },
332
+ "fixed":{
333
+ "p":0.0,
334
+ "r":0.0,
335
+ "f":0.0
336
+ },
337
+ "conj":{
338
+ "p":0.0,
339
+ "r":0.0,
340
+ "f":0.0
341
+ },
342
+ "xcomp":{
343
+ "p":0.0,
344
+ "r":0.0,
345
+ "f":0.0
346
+ },
347
+ "csubj":{
348
+ "p":0.0,
349
+ "r":0.0,
350
+ "f":0.0
351
+ },
352
+ "iobj":{
353
+ "p":0.0,
354
+ "r":0.0,
355
+ "f":0.0
356
+ },
357
+ "vocative":{
358
+ "p":0.0,
359
+ "r":0.0,
360
+ "f":0.0
361
+ },
362
+ "reparandum":{
363
+ "p":0.0,
364
+ "r":0.0,
365
+ "f":0.0
366
+ },
367
+ "discourse":{
368
+ "p":0.0,
369
+ "r":0.0,
370
+ "f":0.0
371
+ },
372
+ "amod":{
373
+ "p":0.0,
374
+ "r":0.0,
375
+ "f":0.0
376
+ },
377
+ "nummod":{
378
+ "p":0.0,
379
+ "r":0.0,
380
+ "f":0.0
381
+ },
382
+ "orphan":{
383
+ "p":0.0,
384
+ "r":0.0,
385
+ "f":0.0
386
+ },
387
+ "flat":{
388
+ "p":0.0,
389
+ "r":0.0,
390
+ "f":0.0
391
+ },
392
+ "obl:npmod":{
393
+ "p":0.0,
394
+ "r":0.0,
395
+ "f":0.0
396
+ },
397
+ "acl":{
398
+ "p":0.0,
399
+ "r":0.0,
400
+ "f":0.0
401
+ }
402
+ },
403
+ "sents_p":0.0059618442,
404
+ "sents_r":0.0694444444,
405
+ "sents_f":0.0109809663,
406
+ "lemma_acc":0.2837307576,
407
+ "tag_acc":0.0,
408
+ "transformer_loss":474.1848888397,
409
+ "parser_loss":897.4665898085,
410
+ "trainable_lemmatizer_loss":137.3459777832,
411
+ "tagger_loss":137.1686668396
412
+ },
413
+ "requirements":[
414
+ "spacy-transformers>=1.3.3,<1.4.0"
415
+ ]
416
+ }
parser/cfg ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "moves":null,
3
+ "update_with_oracle_cut_size":100,
4
+ "multitasks":[
5
+
6
+ ],
7
+ "min_action_freq":30,
8
+ "learn_tokens":false,
9
+ "beam_width":1,
10
+ "beam_density":0.0,
11
+ "beam_update_prob":0.0,
12
+ "incorrect_spans_key":null
13
+ }
parser/model ADDED
Binary file (903 kB). View file
 
parser/moves ADDED
@@ -0,0 +1 @@
 
 
1
+ ��moves�{"0":{"":25539},"1":{"":17115},"2":{"case":6409,"det":5028,"nsubj":3824,"mark":3159,"aux":2722,"cc":1233,"advmod":884,"punct":842,"dislocated":478,"advcl":309,"obl":180,"nummod":76,"vocative":51,"discourse":51,"dep":0},"3":{"obl":3420,"obj":2046,"punct":1929,"nmod":1427,"advmod":1336,"acl:relcl":1051,"conj":949,"advcl":863,"ccomp":575,"parataxis":534,"nsubj":454,"xcomp":444,"det":439,"cop":381,"appos":310,"fixed":236,"dislocated":194,"csubj":103,"vocative":79,"iobj":57,"obl:npmod":52,"flat":40,"dep":0},"4":{"ROOT":1730}}�cfg��neg_key�
tagger/cfg ADDED
@@ -0,0 +1,173 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "label_smoothing":0.0,
3
+ "labels":[
4
+ "AAOR",
5
+ "ACAUS",
6
+ "ACOND",
7
+ "ACOND_PPERS",
8
+ "ACOND_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
9
+ "ACOND_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
10
+ "ACOND_PPERS__Definite=Def|Gender=Masc|Number=Sing|Person=2|PronType=Prs",
11
+ "ACOND_PPERS__Definite=Def|Gender=Masc|Number=Sing|Person=3|PronType=Prs",
12
+ "ACOND_PPERS__Definite=Def|Number=Plur|Person=1|PronType=Prs",
13
+ "ACOND_PPERS__Definite=Def|Number=Plur|Person=2|PronType=Prs",
14
+ "ACOND_PPERS__Definite=Def|Number=Plur|Person=3|PronType=Prs",
15
+ "ACOND_PPERS__Definite=Def|Number=Sing|Person=1|PronType=Prs",
16
+ "ACONJ",
17
+ "ACONJ_PPERS",
18
+ "ACONJ_PPERS__Definite=Def|Number=Sing|Person=1|PronType=Prs",
19
+ "ADV",
20
+ "ADV__Foreign=Yes",
21
+ "AFUTCONJ",
22
+ "AJUS",
23
+ "ALIM",
24
+ "ANEGAOR__Polarity=Neg",
25
+ "ANEGJUS__Polarity=Neg",
26
+ "ANEGOPT__Polarity=Neg",
27
+ "ANEGPST_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|Polarity=Neg|PronType=Prs",
28
+ "ANEGPST__Polarity=Neg",
29
+ "ANY__Polarity=Neg",
30
+ "AOPT",
31
+ "AOPT_PPERS",
32
+ "AOPT_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
33
+ "AOPT_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
34
+ "AOPT_PPERS__Definite=Def|Gender=Masc|Number=Sing|Person=2|PronType=Prs",
35
+ "AOPT_PPERS__Definite=Def|Gender=Masc|Number=Sing|Person=3|PronType=Prs",
36
+ "AOPT_PPERS__Definite=Def|Number=Plur|Person=2|PronType=Prs",
37
+ "AOPT_PPERS__Definite=Def|Number=Plur|Person=3|PronType=Prs",
38
+ "AOPT_PPERS__Definite=Def|Number=Sing|Person=1|PronType=Prs",
39
+ "APREC",
40
+ "APST",
41
+ "APST_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
42
+ "ART",
43
+ "ART__Definite=Def|Gender=Fem|Number=Sing|PronType=Art",
44
+ "ART__Definite=Def|Gender=Masc|Number=Sing|PronType=Art",
45
+ "ART__Definite=Def|Number=Plur|PronType=Art",
46
+ "ART__Definite=Ind|Number=Sing|PronType=Art",
47
+ "ART__PronType=Art",
48
+ "CCIRC",
49
+ "CCIRC_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
50
+ "CCIRC_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
51
+ "CFOC",
52
+ "CFOC_PPERS",
53
+ "CFOC_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
54
+ "CONJ",
55
+ "CONJ__Foreign=Yes",
56
+ "CONJ__Foreign=Yes|Polarity=Neg",
57
+ "COP__Gender=Fem|Number=Sing",
58
+ "COP__Gender=Masc|Number=Sing",
59
+ "COP__Number=Plur",
60
+ "CPRET",
61
+ "CPRET_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
62
+ "CREL",
63
+ "CREL_PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
64
+ "EXIST",
65
+ "EXIST__Foreign=Yes",
66
+ "EXIST__Polarity=Neg",
67
+ "FM",
68
+ "FM__Foreign=Yes",
69
+ "FM__Foreign=Yes|Polarity=Neg",
70
+ "FM__Polarity=Neg",
71
+ "FUT",
72
+ "IMOD",
73
+ "IMOD_PPERO__Definite=Def|Number=Sing|Person=1|PronType=Prs",
74
+ "IMOD__Definite=Def|PronType=Prs|Reflex=Yes",
75
+ "IMOD__PronType=Tot",
76
+ "N",
77
+ "NEG__Foreign=Yes|Polarity=Neg",
78
+ "NEG__Polarity=Neg",
79
+ "NPROP",
80
+ "NPROP__Foreign=Yes",
81
+ "NUM__Foreign=Yes|NumType=Card",
82
+ "NUM__NumType=Card",
83
+ "N_PPERO__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
84
+ "N_PPERO__Definite=Def|Number=Sing|Person=1|PronType=Prs",
85
+ "N__Foreign=Yes",
86
+ "N__PronType=Rcp",
87
+ "PDEM",
88
+ "PDEM__Definite=Def|Foreign=Yes|Number=Plur|PronType=Dem",
89
+ "PDEM__Definite=Def|Gender=Fem|Number=Sing|PronType=Dem",
90
+ "PDEM__Definite=Def|Gender=Masc|Number=Sing|PronType=Dem",
91
+ "PDEM__Definite=Def|Number=Plur|PronType=Dem",
92
+ "PINT__PronType=Ind",
93
+ "PINT__PronType=Int",
94
+ "PPERI__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
95
+ "PPERI__Definite=Def|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
96
+ "PPERI__Definite=Def|Gender=Masc|Number=Sing|Person=2|PronType=Prs",
97
+ "PPERI__Definite=Def|Gender=Masc|Number=Sing|Person=3|PronType=Prs",
98
+ "PPERI__Definite=Def|Number=Plur|Person=1|PronType=Prs",
99
+ "PPERI__Definite=Def|Number=Plur|Person=2|PronType=Prs",
100
+ "PPERI__Definite=Def|Number=Plur|Person=3|PronType=Prs",
101
+ "PPERI__Definite=Def|Number=Sing|Person=1|PronType=Prs",
102
+ "PPERO",
103
+ "PPERO__Definite=Def|Foreign=Yes|Number=Plur|Person=3|PronType=Prs",
104
+ "PPERO__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
105
+ "PPERO__Definite=Def|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
106
+ "PPERO__Definite=Def|Gender=Masc|Number=Sing|Person=2|PronType=Prs",
107
+ "PPERO__Definite=Def|Gender=Masc|Number=Sing|Person=3|PronType=Prs",
108
+ "PPERO__Definite=Def|Number=Plur|Person=1|PronType=Prs",
109
+ "PPERO__Definite=Def|Number=Plur|Person=2|PronType=Prs",
110
+ "PPERO__Definite=Def|Number=Plur|Person=3|PronType=Prs",
111
+ "PPERO__Definite=Def|Number=Sing|Person=1|PronType=Prs",
112
+ "PPERO__Definite=Def|Poss=Yes|PronType=Prs",
113
+ "PPERS",
114
+ "PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
115
+ "PPERS__Definite=Def|Gender=Fem|Number=Sing|Person=3|PronType=Prs",
116
+ "PPERS__Definite=Def|Gender=Masc|Number=Sing|Person=2|PronType=Prs",
117
+ "PPERS__Definite=Def|Gender=Masc|Number=Sing|Person=3|PronType=Prs",
118
+ "PPERS__Definite=Def|Number=Plur|Person=1|PronType=Prs",
119
+ "PPERS__Definite=Def|Number=Plur|Person=2|PronType=Prs",
120
+ "PPERS__Definite=Def|Number=Plur|Person=3|PronType=Prs",
121
+ "PPERS__Definite=Def|Number=Sing|Person=1|PronType=Prs",
122
+ "PPOS",
123
+ "PPOS__Definite=Def|Gender=Fem|Gender[psor]=Fem|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs",
124
+ "PPOS__Definite=Def|Gender=Fem|Gender[psor]=Fem|Number=Sing|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs",
125
+ "PPOS__Definite=Def|Gender=Fem|Gender[psor]=Masc|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs",
126
+ "PPOS__Definite=Def|Gender=Fem|Gender[psor]=Masc|Number=Sing|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs",
127
+ "PPOS__Definite=Def|Gender=Fem|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs",
128
+ "PPOS__Definite=Def|Gender=Fem|Number=Sing|Number[psor]=Plur|Person=2|Poss=Yes|PronType=Prs",
129
+ "PPOS__Definite=Def|Gender=Fem|Number=Sing|Number[psor]=Plur|Person=3|Poss=Yes|PronType=Prs",
130
+ "PPOS__Definite=Def|Gender=Fem|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs",
131
+ "PPOS__Definite=Def|Gender=Fem|Number=Sing|Poss=Yes|PronType=Prs",
132
+ "PPOS__Definite=Def|Gender=Masc|Gender[psor]=Fem|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs",
133
+ "PPOS__Definite=Def|Gender=Masc|Gender[psor]=Fem|Number=Sing|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs",
134
+ "PPOS__Definite=Def|Gender=Masc|Gender[psor]=Masc|Number=Sing|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs",
135
+ "PPOS__Definite=Def|Gender=Masc|Gender[psor]=Masc|Number=Sing|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs",
136
+ "PPOS__Definite=Def|Gender=Masc|Number=Sing|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs",
137
+ "PPOS__Definite=Def|Gender=Masc|Number=Sing|Number[psor]=Plur|Person=2|Poss=Yes|PronType=Prs",
138
+ "PPOS__Definite=Def|Gender=Masc|Number=Sing|Number[psor]=Plur|Person=3|Poss=Yes|PronType=Prs",
139
+ "PPOS__Definite=Def|Gender=Masc|Number=Sing|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs",
140
+ "PPOS__Definite=Def|Gender=Masc|Number=Sing|Poss=Yes|PronType=Prs",
141
+ "PPOS__Definite=Def|Gender[psor]=Fem|Number=Plur|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs",
142
+ "PPOS__Definite=Def|Gender[psor]=Fem|Number=Plur|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs",
143
+ "PPOS__Definite=Def|Gender[psor]=Masc|Number=Plur|Number[psor]=Sing|Person=2|Poss=Yes|PronType=Prs",
144
+ "PPOS__Definite=Def|Gender[psor]=Masc|Number=Plur|Number[psor]=Sing|Person=3|Poss=Yes|PronType=Prs",
145
+ "PPOS__Definite=Def|Number=Plur|Number[psor]=Plur|Person=1|Poss=Yes|PronType=Prs",
146
+ "PPOS__Definite=Def|Number=Plur|Number[psor]=Plur|Person=2|Poss=Yes|PronType=Prs",
147
+ "PPOS__Definite=Def|Number=Plur|Number[psor]=Plur|Person=3|Poss=Yes|PronType=Prs",
148
+ "PPOS__Definite=Def|Number=Plur|Number[psor]=Sing|Person=1|Poss=Yes|PronType=Prs",
149
+ "PPOS__Definite=Def|Number=Plur|Poss=Yes|PronType=Prs",
150
+ "PREP",
151
+ "PREP_PPERO",
152
+ "PREP_PPERO__Definite=Def|Gender=Fem|Number=Sing|Person=2|PronType=Prs",
153
+ "PREP_PPERO__Definite=Def|Number=Sing|Person=1|PronType=Prs",
154
+ "PREP__Foreign=Yes",
155
+ "PTC",
156
+ "PTC__Foreign=Yes",
157
+ "PTC__Polarity=Neg",
158
+ "PUNCT",
159
+ "UNKNOWN",
160
+ "UNKNOWN__Foreign=Yes",
161
+ "VBD__VerbForm=Fin",
162
+ "VIMP__Foreign=Yes|VerbForm=Fin",
163
+ "VIMP__VerbForm=Fin",
164
+ "VSTAT__VerbForm=Fin",
165
+ "V_PPERO__VerbForm=Fin",
166
+ "V__Foreign=Yes|VerbForm=Fin",
167
+ "V__Foreign=Yes|VerbForm=Inf",
168
+ "V__VerbForm=Fin",
169
+ "V__VerbForm=Inf"
170
+ ],
171
+ "neg_prefix":"!",
172
+ "overwrite":false
173
+ }
tagger/model ADDED
Binary file (511 kB). View file
 
tokenizer ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ��prefix_search� �^§|^%|^=|^—|^–|^\+(?![0-9])|^…|^……|^,|^:|^;|^\!|^\?|^¿|^؟|^¡|^\(|^\)|^\[|^\]|^\{|^\}|^<|^>|^_|^#|^\*|^&|^。|^？|^！|^，|^、|^；|^：|^～|^·|^।|^،|^۔|^؛|^٪|^\.\.+|^…|^\'|^"|^”|^“|^`|^‘|^´|^’|^‚|^,|^„|^»|^«|^「|^」|^『|^』|^（|^）|^〔|^〕|^【|^】|^《|^》|^〈|^〉|^〈|^〉|^⟦|^⟧|^\$|^£|^€|^¥|^฿|^US\$|^C\$|^A\$|^₽|^﷼|^₴|^₠|^₡|^₢|^₣|^₤|^₥|^₦|^₧|^₨|^₩|^₪|^₫|^€|^₭|^₮|^₯|^₰|^₱|^₲|^₳|^₴|^₵|^₶|^₷|^₸|^₹|^₺|^₻|^₼|^₽|^₾|^₿|^[\u00A6\u00A9\u00AE\u00B0\u0482\u058D\u058E\u060E\u060F\u06DE\u06E9\u06FD\u06FE\u07F6\u09FA\u0B70\u0BF3-\u0BF8\u0BFA\u0C7F\u0D4F\u0D79\u0F01-\u0F03\u0F13\u0F15-\u0F17\u0F1A-\u0F1F\u0F34\u0F36\u0F38\u0FBE-\u0FC5\u0FC7-\u0FCC\u0FCE\u0FCF\u0FD5-\u0FD8\u109E\u109F\u1390-\u1399\u1940\u19DE-\u19FF\u1B61-\u1B6A\u1B74-\u1B7C\u2100\u2101\u2103-\u2106\u2108\u2109\u2114\u2116\u2117\u211E-\u2123\u2125\u2127\u2129\u212E\u213A\u213B\u214A\u214C\u214D\u214F\u218A\u218B\u2195-\u2199\u219C-\u219F\u21A1\u21A2\u21A4\u21A5\u21A7-\u21AD\u21AF-\u21CD\u21D0\u21D1\u21D3\u21D5-\u21F3\u2300-\u2307\u230C-\u231F\u2322-\u2328\u232B-\u237B\u237D-\u239A\u23B4-\u23DB\u23E2-\u2426\u2440-\u244A\u249C-\u24E9\u2500-\u25B6\u25B8-\u25C0\u25C2-\u25F7\u2600-\u266E\u2670-\u2767\u2794-\u27BF\u2800-\u28FF\u2B00-\u2B2F\u2B45\u2B46\u2B4D-\u2B73\u2B76-\u2B95\u2B98-\u2BC8\u2BCA-\u2BFE\u2CE5-\u2CEA\u2E80-\u2E99\u2E9B-\u2EF3\u2F00-\u2FD5\u2FF0-\u2FFB\u3004\u3012\u3013\u3020\u3036\u3037\u303E\u303F\u3190\u3191\u3196-\u319F\u31C0-\u31E3\u3200-\u321E\u322A-\u3247\u3250\u3260-\u327F\u328A-\u32B0\u32C0-\u32FE\u3300-\u33FF\u4DC0-\u4DFF\uA490-\uA4C6\uA828-\uA82B\uA836\uA837\uA839\uAA77-\uAA79\uFDFD\uFFE4\uFFE8\uFFED\uFFEE\uFFFC\uFFFD\U00010137-\U0001013F\U00010179-\U00010189\U0001018C-\U0001018E\U00010190-\U0001019B\U000101A0\U000101D0-\U000101FC\U00010877\U00010878\U00010AC8\U0001173F\U00016B3C-\U00016B3F\U00016B45\U0001BC9C\U0001D000-\U0001D0F5\U0001D100-\U0001D126\U0001D129-\U0001D164\U0001D16A-\U0001D16C\U0001D183\U0001D184\U0001D18C-\U0001D1A9\U0001D1AE-\U0001D1E8\U0001D200-\U0001D241\U0001D245\U0001D300-\U0001D356\U0001D800-\U0001D9FF\U0001DA37-\U0001DA3A\U0001DA6D-\U0001DA74\U0001DA76-\U0001DA83\U0001DA85\U0001DA86\U0001ECAC\U0001F000-\U0001F02B\U0001F030-\U0001F093\U0001F0A0-\U0001F0AE\U0001F0B1-\U0001F0BF\U0001F0C1-\U0001F0CF\U0001F0D1-\U0001F0F5\U0001F110-\U0001F16B\U0001F170-\U0001F1AC\U0001F1E6-\U0001F202\U0001F210-\U0001F23B\U0001F240-\U0001F248\U0001F250\U0001F251\U0001F260-\U0001F265\U0001F300-\U0001F3FA\U0001F400-\U0001F6D4\U0001F6E0-\U0001F6EC\U0001F6F0-\U0001F6F9\U0001F700-\U0001F773\U0001F780-\U0001F7D8\U0001F800-\U0001F80B\U0001F810-\U0001F847\U0001F850-\U0001F859\U0001F860-\U0001F887\U0001F890-\U0001F8AD\U0001F900-\U0001F90B\U0001F910-\U0001F93E\U0001F940-\U0001F970\U0001F973-\U0001F976\U0001F97A\U0001F97C-\U0001F9A2\U0001F9B0-\U0001F9B9\U0001F9C0-\U0001F9C2\U0001F9D0-\U0001F9FF\U0001FA60-\U0001FA6D]�suffix_search�2�…$|……$|,$|:$|;$|\!$|\?$|¿$|؟$|¡$|\($|\)$|\[$|\]$|\{$|\}$|<$|>$|_$|#$|\*$|&$|。$|？$|！$|，$|、$|；$|：$|～$|·$|।$|،$|۔$|؛$|٪$|\.\.+$|…$|\'$|"$|”$|“$|`$|‘$|´$|’$|‚$|,$|„$|»$|«$|「$|」$|『$|』$|（$|）$|〔$|〕$|【$|】$|《$|》$|〈$|〉$|〈$|〉$|⟦$|⟧$|[\u00A6\u00A9\u00AE\u00B0\u0482\u058D\u058E\u060E\u060F\u06DE\u06E9\u06FD\u06FE\u07F6\u09FA\u0B70\u0BF3-\u0BF8\u0BFA\u0C7F\u0D4F\u0D79\u0F01-\u0F03\u0F13\u0F15-\u0F17\u0F1A-\u0F1F\u0F34\u0F36\u0F38\u0FBE-\u0FC5\u0FC7-\u0FCC\u0FCE\u0FCF\u0FD5-\u0FD8\u109E\u109F\u1390-\u1399\u1940\u19DE-\u19FF\u1B61-\u1B6A\u1B74-\u1B7C\u2100\u2101\u2103-\u2106\u2108\u2109\u2114\u2116\u2117\u211E-\u2123\u2125\u2127\u2129\u212E\u213A\u213B\u214A\u214C\u214D\u214F\u218A\u218B\u2195-\u2199\u219C-\u219F\u21A1\u21A2\u21A4\u21A5\u21A7-\u21AD\u21AF-\u21CD\u21D0\u21D1\u21D3\u21D5-\u21F3\u2300-\u2307\u230C-\u231F\u2322-\u2328\u232B-\u237B\u237D-\u239A\u23B4-\u23DB\u23E2-\u2426\u2440-\u244A\u249C-\u24E9\u2500-\u25B6\u25B8-\u25C0\u25C2-\u25F7\u2600-\u266E\u2670-\u2767\u2794-\u27BF\u2800-\u28FF\u2B00-\u2B2F\u2B45\u2B46\u2B4D-\u2B73\u2B76-\u2B95\u2B98-\u2BC8\u2BCA-\u2BFE\u2CE5-\u2CEA\u2E80-\u2E99\u2E9B-\u2EF3\u2F00-\u2FD5\u2FF0-\u2FFB\u3004\u3012\u3013\u3020\u3036\u3037\u303E\u303F\u3190\u3191\u3196-\u319F\u31C0-\u31E3\u3200-\u321E\u322A-\u3247\u3250\u3260-\u327F\u328A-\u32B0\u32C0-\u32FE\u3300-\u33FF\u4DC0-\u4DFF\uA490-\uA4C6\uA828-\uA82B\uA836\uA837\uA839\uAA77-\uAA79\uFDFD\uFFE4\uFFE8\uFFED\uFFEE\uFFFC\uFFFD\U00010137-\U0001013F\U00010179-\U00010189\U0001018C-\U0001018E\U00010190-\U0001019B\U000101A0\U000101D0-\U000101FC\U00010877\U00010878\U00010AC8\U0001173F\U00016B3C-\U00016B3F\U00016B45\U0001BC9C\U0001D000-\U0001D0F5\U0001D100-\U0001D126\U0001D129-\U0001D164\U0001D16A-\U0001D16C\U0001D183\U0001D184\U0001D18C-\U0001D1A9\U0001D1AE-\U0001D1E8\U0001D200-\U0001D241\U0001D245\U0001D300-\U0001D356\U0001D800-\U0001D9FF\U0001DA37-\U0001DA3A\U0001DA6D-\U0001DA74\U0001DA76-\U0001DA83\U0001DA85\U0001DA86\U0001ECAC\U0001F000-\U0001F02B\U0001F030-\U0001F093\U0001F0A0-\U0001F0AE\U0001F0B1-\U0001F0BF\U0001F0C1-\U0001F0CF\U0001F0D1-\U0001F0F5\U0001F110-\U0001F16B\U0001F170-\U0001F1AC\U0001F1E6-\U0001F202\U0001F210-\U0001F23B\U0001F240-\U0001F248\U0001F250\U0001F251\U0001F260-\U0001F265\U0001F300-\U0001F3FA\U0001F400-\U0001F6D4\U0001F6E0-\U0001F6EC\U0001F6F0-\U0001F6F9\U0001F700-\U0001F773\U0001F780-\U0001F7D8\U0001F800-\U0001F80B\U0001F810-\U0001F847\U0001F850-\U0001F859\U0001F860-\U0001F887\U0001F890-\U0001F8AD\U0001F900-\U0001F90B\U0001F910-\U0001F93E\U0001F940-\U0001F970\U0001F973-\U0001F976\U0001F97A\U0001F97C-\U0001F9A2\U0001F9B0-\U0001F9B9\U0001F9C0-\U0001F9C2\U0001F9D0-\U0001F9FF\U0001FA60-\U0001FA6D]$|'s$|'S$|’s$|’S$|—$|–$|(?<=[0-9])\+$|(?<=°[FfCcKk])\.$|(?<=[0-9])(?:\$|£|€|¥|฿|US\$|C\$|A\$|₽|﷼|₴|₠|₡|₢|₣|₤|₥|₦|₧|₨|₩|₪|₫|€|₭|₮|₯|₰|₱|₲|₳|₴|₵|₶|₷|₸|₹|₺|₻|₼|₽|₾|₿)$|(?<=[0-9])(?:km|km²|km³|m|m²|m³|dm|dm²|dm³|cm|cm²|cm³|mm|mm²|mm³|ha|µm|nm|yd|in|ft|kg|g|mg|µg|t|lb|oz|m/s|km/h|kmh|mph|hPa|Pa|mbar|mb|MB|kb|KB|gb|GB|tb|TB|T|G|M|K|%|км|км²|км³|м|м²|м³|дм|дм²|дм³|см|см²|см³|мм|мм²|мм³|нм|кг|г|мг|м/с|км/ч|кПа|Па|мбар|Кб|КБ|кб|Мб|МБ|мб|Гб|ГБ|гб|Тб|ТБ|тбكم|كم²|كم³|م|م²|م³|سم|سم²|سم³|مم|مم²|مم³|كم|غرام|جرام|جم|كغ|ملغ|كوب|اكواب)$|(?<=[0-9a-z\uFF41-\uFF5A\u00DF-\u00F6\u00F8-\u00FF\u0101\u0103\u0105\u0107\u0109\u010B\u010D\u010F\u0111\u0113\u0115\u0117\u0119\u011B\u011D\u011F\u0121\u0123\u0125\u0127\u0129\u012B\u012D\u012F\u0131\u0133\u0135\u0137\u0138\u013A\u013C\u013E\u0140\u0142\u0144\u0146\u0148\u0149\u014B\u014D\u014F\u0151\u0153\u0155\u0157\u0159\u015B\u015D\u015F\u0161\u0163\u0165\u0167\u0169\u016B\u016D\u016F\u0171\u0173\u0175\u0177\u017A\u017C\u017E\u017F\u0180\u0183\u0185\u0188\u018C\u018D\u0192\u0195\u0199-\u019B\u019E\u01A1\u01A3\u01A5\u01A8\u01AA\u01AB\u01AD\u01B0\u01B4\u01B6\u01B9\u01BA\u01BD-\u01BF\u01C6\u01C9\u01CC\u01CE\u01D0\u01D2\u01D4\u01D6\u01D8\u01DA\u01DC\u01DD\u01DF\u01E1\u01E3\u01E5\u01E7\u01E9\u01EB\u01ED\u01EF\u01F0\u01F3\u01F5\u01F9\u01FB\u01FD\u01FF\u0201\u0203\u0205\u0207\u0209\u020B\u020D\u020F\u0211\u0213\u0215\u0217\u0219\u021B\u021D\u021F\u0221\u0223\u0225\u0227\u0229\u022B\u022D\u022F\u0231\u0233-\u0239\u023C\u023F\u0240\u0242\u0247\u0249\u024B\u024D\u024F\u2C61\u2C65\u2C66\u2C68\u2C6A\u2C6C\u2C71\u2C73\u2C74\u2C76-\u2C7B\uA723\uA725\uA727\uA729\uA72B\uA72D\uA72F-\uA731\uA733\uA735\uA737\uA739\uA73B\uA73D\uA73F\uA741\uA743\uA745\uA747\uA749\uA74B\uA74D\uA74F\uA751\uA753\uA755\uA757\uA759\uA75B\uA75D\uA75F\uA761\uA763\uA765\uA767\uA769\uA76B\uA76D\uA76F\uA771-\uA778\uA77A\uA77C\uA77F\uA781\uA783\uA785\uA787\uA78C\uA78E\uA791\uA793-\uA795\uA797\uA799\uA79B\uA79D\uA79F\uA7A1\uA7A3\uA7A5\uA7A7\uA7A9\uA7AF\uA7B5\uA7B7\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E01\u1E03\u1E05\u1E07\u1E09\u1E0B\u1E0D\u1E0F\u1E11\u1E13\u1E15\u1E17\u1E19\u1E1B\u1E1D\u1E1F\u1E21\u1E23\u1E25\u1E27\u1E29\u1E2B\u1E2D\u1E2F\u1E31\u1E33\u1E35\u1E37\u1E39\u1E3B\u1E3D\u1E3F\u1E41\u1E43\u1E45\u1E47\u1E49\u1E4B\u1E4D\u1E4F\u1E51\u1E53\u1E55\u1E57\u1E59\u1E5B\u1E5D\u1E5F\u1E61\u1E63\u1E65\u1E67\u1E69\u1E6B\u1E6D\u1E6F\u1E71\u1E73\u1E75\u1E77\u1E79\u1E7B\u1E7D\u1E7F\u1E81\u1E83\u1E85\u1E87\u1E89\u1E8B\u1E8D\u1E8F\u1E91\u1E93\u1E95-\u1E9D\u1E9F\u1EA1\u1EA3\u1EA5\u1EA7\u1EA9\u1EAB\u1EAD\u1EAF\u1EB1\u1EB3\u1EB5\u1EB7\u1EB9\u1EBB\u1EBD\u1EBF\u1EC1\u1EC3\u1EC5\u1EC7\u1EC9\u1ECB\u1ECD\u1ECF\u1ED1\u1ED3\u1ED5\u1ED7\u1ED9\u1EDB\u1EDD\u1EDF\u1EE1\u1EE3\u1EE5\u1EE7\u1EE9\u1EEB\u1EED\u1EEF\u1EF1\u1EF3\u1EF5\u1EF7\u1EF9\u1EFB\u1EFD\u1EFFёа-яәөүҗңһα-ωάέίόώήύа-щюяіїєґѓѕјљњќѐѝ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F%²\-\+…|……|,|:|;|\!|\?|¿|؟|¡|\(|\)|\[|\]|\{|\}|<|>|_|#|\*|&|。|？|！|，|、|；|：|～|·|।|،|۔|؛|٪(?:\'"”“`‘´’‚,„»«「」『』（）〔〕【】《》〈〉〈〉⟦⟧)])\.$|(?<=[A-Z\uFF21-\uFF3A\u00C0-\u00D6\u00D8-\u00DE\u0100\u0102\u0104\u0106\u0108\u010A\u010C\u010E\u0110\u0112\u0114\u0116\u0118\u011A\u011C\u011E\u0120\u0122\u0124\u0126\u0128\u012A\u012C\u012E\u0130\u0132\u0134\u0136\u0139\u013B\u013D\u013F\u0141\u0143\u0145\u0147\u014A\u014C\u014E\u0150\u0152\u0154\u0156\u0158\u015A\u015C\u015E\u0160\u0162\u0164\u0166\u0168\u016A\u016C\u016E\u0170\u0172\u0174\u0176\u0178\u0179\u017B\u017D\u0181\u0182\u0184\u0186\u0187\u0189-\u018B\u018E-\u0191\u0193\u0194\u0196-\u0198\u019C\u019D\u019F\u01A0\u01A2\u01A4\u01A6\u01A7\u01A9\u01AC\u01AE\u01AF\u01B1-\u01B3\u01B5\u01B7\u01B8\u01BC\u01C4\u01C7\u01CA\u01CD\u01CF\u01D1\u01D3\u01D5\u01D7\u01D9\u01DB\u01DE\u01E0\u01E2\u01E4\u01E6\u01E8\u01EA\u01EC\u01EE\u01F1\u01F4\u01F6-\u01F8\u01FA\u01FC\u01FE\u0200\u0202\u0204\u0206\u0208\u020A\u020C\u020E\u0210\u0212\u0214\u0216\u0218\u021A\u021C\u021E\u0220\u0222\u0224\u0226\u0228\u022A\u022C\u022E\u0230\u0232\u023A\u023B\u023D\u023E\u0241\u0243-\u0246\u0248\u024A\u024C\u024E\u2C60\u2C62-\u2C64\u2C67\u2C69\u2C6B\u2C6D-\u2C70\u2C72\u2C75\u2C7E\u2C7F\uA722\uA724\uA726\uA728\uA72A\uA72C\uA72E\uA732\uA734\uA736\uA738\uA73A\uA73C\uA73E\uA740\uA742\uA744\uA746\uA748\uA74A\uA74C\uA74E\uA750\uA752\uA754\uA756\uA758\uA75A\uA75C\uA75E\uA760\uA762\uA764\uA766\uA768\uA76A\uA76C\uA76E\uA779\uA77B\uA77D\uA77E\uA780\uA782\uA784\uA786\uA78B\uA78D\uA790\uA792\uA796\uA798\uA79A\uA79C\uA79E\uA7A0\uA7A2\uA7A4\uA7A6\uA7A8\uA7AA-\uA7AE\uA7B0-\uA7B4\uA7B6\uA7B8\u1E00\u1E02\u1E04\u1E06\u1E08\u1E0A\u1E0C\u1E0E\u1E10\u1E12\u1E14\u1E16\u1E18\u1E1A\u1E1C\u1E1E\u1E20\u1E22\u1E24\u1E26\u1E28\u1E2A\u1E2C\u1E2E\u1E30\u1E32\u1E34\u1E36\u1E38\u1E3A\u1E3C\u1E3E\u1E40\u1E42\u1E44\u1E46\u1E48\u1E4A\u1E4C\u1E4E\u1E50\u1E52\u1E54\u1E56\u1E58\u1E5A\u1E5C\u1E5E\u1E60\u1E62\u1E64\u1E66\u1E68\u1E6A\u1E6C\u1E6E\u1E70\u1E72\u1E74\u1E76\u1E78\u1E7A\u1E7C\u1E7E\u1E80\u1E82\u1E84\u1E86\u1E88\u1E8A\u1E8C\u1E8E\u1E90\u1E92\u1E94\u1E9E\u1EA0\u1EA2\u1EA4\u1EA6\u1EA8\u1EAA\u1EAC\u1EAE\u1EB0\u1EB2\u1EB4\u1EB6\u1EB8\u1EBA\u1EBC\u1EBE\u1EC0\u1EC2\u1EC4\u1EC6\u1EC8\u1ECA\u1ECC\u1ECE\u1ED0\u1ED2\u1ED4\u1ED6\u1ED8\u1EDA\u1EDC\u1EDE\u1EE0\u1EE2\u1EE4\u1EE6\u1EE8\u1EEA\u1EEC\u1EEE\u1EF0\u1EF2\u1EF4\u1EF6\u1EF8\u1EFA\u1EFC\u1EFEЁА-ЯӘӨҮҖҢҺΑ-ΩΆΈΊΌΏΉΎА-ЩЮЯІЇЄҐЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F][A-Z\uFF21-\uFF3A\u00C0-\u00D6\u00D8-\u00DE\u0100\u0102\u0104\u0106\u0108\u010A\u010C\u010E\u0110\u0112\u0114\u0116\u0118\u011A\u011C\u011E\u0120\u0122\u0124\u0126\u0128\u012A\u012C\u012E\u0130\u0132\u0134\u0136\u0139\u013B\u013D\u013F\u0141\u0143\u0145\u0147\u014A\u014C\u014E\u0150\u0152\u0154\u0156\u0158\u015A\u015C\u015E\u0160\u0162\u0164\u0166\u0168\u016A\u016C\u016E\u0170\u0172\u0174\u0176\u0178\u0179\u017B\u017D\u0181\u0182\u0184\u0186\u0187\u0189-\u018B\u018E-\u0191\u0193\u0194\u0196-\u0198\u019C\u019D\u019F\u01A0\u01A2\u01A4\u01A6\u01A7\u01A9\u01AC\u01AE\u01AF\u01B1-\u01B3\u01B5\u01B7\u01B8\u01BC\u01C4\u01C7\u01CA\u01CD\u01CF\u01D1\u01D3\u01D5\u01D7\u01D9\u01DB\u01DE\u01E0\u01E2\u01E4\u01E6\u01E8\u01EA\u01EC\u01EE\u01F1\u01F4\u01F6-\u01F8\u01FA\u01FC\u01FE\u0200\u0202\u0204\u0206\u0208\u020A\u020C\u020E\u0210\u0212\u0214\u0216\u0218\u021A\u021C\u021E\u0220\u0222\u0224\u0226\u0228\u022A\u022C\u022E\u0230\u0232\u023A\u023B\u023D\u023E\u0241\u0243-\u0246\u0248\u024A\u024C\u024E\u2C60\u2C62-\u2C64\u2C67\u2C69\u2C6B\u2C6D-\u2C70\u2C72\u2C75\u2C7E\u2C7F\uA722\uA724\uA726\uA728\uA72A\uA72C\uA72E\uA732\uA734\uA736\uA738\uA73A\uA73C\uA73E\uA740\uA742\uA744\uA746\uA748\uA74A\uA74C\uA74E\uA750\uA752\uA754\uA756\uA758\uA75A\uA75C\uA75E\uA760\uA762\uA764\uA766\uA768\uA76A\uA76C\uA76E\uA779\uA77B\uA77D\uA77E\uA780\uA782\uA784\uA786\uA78B\uA78D\uA790\uA792\uA796\uA798\uA79A\uA79C\uA79E\uA7A0\uA7A2\uA7A4\uA7A6\uA7A8\uA7AA-\uA7AE\uA7B0-\uA7B4\uA7B6\uA7B8\u1E00\u1E02\u1E04\u1E06\u1E08\u1E0A\u1E0C\u1E0E\u1E10\u1E12\u1E14\u1E16\u1E18\u1E1A\u1E1C\u1E1E\u1E20\u1E22\u1E24\u1E26\u1E28\u1E2A\u1E2C\u1E2E\u1E30\u1E32\u1E34\u1E36\u1E38\u1E3A\u1E3C\u1E3E\u1E40\u1E42\u1E44\u1E46\u1E48\u1E4A\u1E4C\u1E4E\u1E50\u1E52\u1E54\u1E56\u1E58\u1E5A\u1E5C\u1E5E\u1E60\u1E62\u1E64\u1E66\u1E68\u1E6A\u1E6C\u1E6E\u1E70\u1E72\u1E74\u1E76\u1E78\u1E7A\u1E7C\u1E7E\u1E80\u1E82\u1E84\u1E86\u1E88\u1E8A\u1E8C\u1E8E\u1E90\u1E92\u1E94\u1E9E\u1EA0\u1EA2\u1EA4\u1EA6\u1EA8\u1EAA\u1EAC\u1EAE\u1EB0\u1EB2\u1EB4\u1EB6\u1EB8\u1EBA\u1EBC\u1EBE\u1EC0\u1EC2\u1EC4\u1EC6\u1EC8\u1ECA\u1ECC\u1ECE\u1ED0\u1ED2\u1ED4\u1ED6\u1ED8\u1EDA\u1EDC\u1EDE\u1EE0\u1EE2\u1EE4\u1EE6\u1EE8\u1EEA\u1EEC\u1EEE\u1EF0\u1EF2\u1EF4\u1EF6\u1EF8\u1EFA\u1EFC\u1EFEЁА-ЯӘӨҮҖҢҺΑ-ΩΆΈΊΌΏΉΎА-ЩЮЯІЇЄҐЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])\.$�infix_finditer�>�\.\.+|…|[\u00A6\u00A9\u00AE\u00B0\u0482\u058D\u058E\u060E\u060F\u06DE\u06E9\u06FD\u06FE\u07F6\u09FA\u0B70\u0BF3-\u0BF8\u0BFA\u0C7F\u0D4F\u0D79\u0F01-\u0F03\u0F13\u0F15-\u0F17\u0F1A-\u0F1F\u0F34\u0F36\u0F38\u0FBE-\u0FC5\u0FC7-\u0FCC\u0FCE\u0FCF\u0FD5-\u0FD8\u109E\u109F\u1390-\u1399\u1940\u19DE-\u19FF\u1B61-\u1B6A\u1B74-\u1B7C\u2100\u2101\u2103-\u2106\u2108\u2109\u2114\u2116\u2117\u211E-\u2123\u2125\u2127\u2129\u212E\u213A\u213B\u214A\u214C\u214D\u214F\u218A\u218B\u2195-\u2199\u219C-\u219F\u21A1\u21A2\u21A4\u21A5\u21A7-\u21AD\u21AF-\u21CD\u21D0\u21D1\u21D3\u21D5-\u21F3\u2300-\u2307\u230C-\u231F\u2322-\u2328\u232B-\u237B\u237D-\u239A\u23B4-\u23DB\u23E2-\u2426\u2440-\u244A\u249C-\u24E9\u2500-\u25B6\u25B8-\u25C0\u25C2-\u25F7\u2600-\u266E\u2670-\u2767\u2794-\u27BF\u2800-\u28FF\u2B00-\u2B2F\u2B45\u2B46\u2B4D-\u2B73\u2B76-\u2B95\u2B98-\u2BC8\u2BCA-\u2BFE\u2CE5-\u2CEA\u2E80-\u2E99\u2E9B-\u2EF3\u2F00-\u2FD5\u2FF0-\u2FFB\u3004\u3012\u3013\u3020\u3036\u3037\u303E\u303F\u3190\u3191\u3196-\u319F\u31C0-\u31E3\u3200-\u321E\u322A-\u3247\u3250\u3260-\u327F\u328A-\u32B0\u32C0-\u32FE\u3300-\u33FF\u4DC0-\u4DFF\uA490-\uA4C6\uA828-\uA82B\uA836\uA837\uA839\uAA77-\uAA79\uFDFD\uFFE4\uFFE8\uFFED\uFFEE\uFFFC\uFFFD\U00010137-\U0001013F\U00010179-\U00010189\U0001018C-\U0001018E\U00010190-\U0001019B\U000101A0\U000101D0-\U000101FC\U00010877\U00010878\U00010AC8\U0001173F\U00016B3C-\U00016B3F\U00016B45\U0001BC9C\U0001D000-\U0001D0F5\U0001D100-\U0001D126\U0001D129-\U0001D164\U0001D16A-\U0001D16C\U0001D183\U0001D184\U0001D18C-\U0001D1A9\U0001D1AE-\U0001D1E8\U0001D200-\U0001D241\U0001D245\U0001D300-\U0001D356\U0001D800-\U0001D9FF\U0001DA37-\U0001DA3A\U0001DA6D-\U0001DA74\U0001DA76-\U0001DA83\U0001DA85\U0001DA86\U0001ECAC\U0001F000-\U0001F02B\U0001F030-\U0001F093\U0001F0A0-\U0001F0AE\U0001F0B1-\U0001F0BF\U0001F0C1-\U0001F0CF\U0001F0D1-\U0001F0F5\U0001F110-\U0001F16B\U0001F170-\U0001F1AC\U0001F1E6-\U0001F202\U0001F210-\U0001F23B\U0001F240-\U0001F248\U0001F250\U0001F251\U0001F260-\U0001F265\U0001F300-\U0001F3FA\U0001F400-\U0001F6D4\U0001F6E0-\U0001F6EC\U0001F6F0-\U0001F6F9\U0001F700-\U0001F773\U0001F780-\U0001F7D8\U0001F800-\U0001F80B\U0001F810-\U0001F847\U0001F850-\U0001F859\U0001F860-\U0001F887\U0001F890-\U0001F8AD\U0001F900-\U0001F90B\U0001F910-\U0001F93E\U0001F940-\U0001F970\U0001F973-\U0001F976\U0001F97A\U0001F97C-\U0001F9A2\U0001F9B0-\U0001F9B9\U0001F9C0-\U0001F9C2\U0001F9D0-\U0001F9FF\U0001FA60-\U0001FA6D]|(?<=[0-9])[+\-\*^](?=[0-9-])|(?<=[a-z\uFF41-\uFF5A\u00DF-\u00F6\u00F8-\u00FF\u0101\u0103\u0105\u0107\u0109\u010B\u010D\u010F\u0111\u0113\u0115\u0117\u0119\u011B\u011D\u011F\u0121\u0123\u0125\u0127\u0129\u012B\u012D\u012F\u0131\u0133\u0135\u0137\u0138\u013A\u013C\u013E\u0140\u0142\u0144\u0146\u0148\u0149\u014B\u014D\u014F\u0151\u0153\u0155\u0157\u0159\u015B\u015D\u015F\u0161\u0163\u0165\u0167\u0169\u016B\u016D\u016F\u0171\u0173\u0175\u0177\u017A\u017C\u017E\u017F\u0180\u0183\u0185\u0188\u018C\u018D\u0192\u0195\u0199-\u019B\u019E\u01A1\u01A3\u01A5\u01A8\u01AA\u01AB\u01AD\u01B0\u01B4\u01B6\u01B9\u01BA\u01BD-\u01BF\u01C6\u01C9\u01CC\u01CE\u01D0\u01D2\u01D4\u01D6\u01D8\u01DA\u01DC\u01DD\u01DF\u01E1\u01E3\u01E5\u01E7\u01E9\u01EB\u01ED\u01EF\u01F0\u01F3\u01F5\u01F9\u01FB\u01FD\u01FF\u0201\u0203\u0205\u0207\u0209\u020B\u020D\u020F\u0211\u0213\u0215\u0217\u0219\u021B\u021D\u021F\u0221\u0223\u0225\u0227\u0229\u022B\u022D\u022F\u0231\u0233-\u0239\u023C\u023F\u0240\u0242\u0247\u0249\u024B\u024D\u024F\u2C61\u2C65\u2C66\u2C68\u2C6A\u2C6C\u2C71\u2C73\u2C74\u2C76-\u2C7B\uA723\uA725\uA727\uA729\uA72B\uA72D\uA72F-\uA731\uA733\uA735\uA737\uA739\uA73B\uA73D\uA73F\uA741\uA743\uA745\uA747\uA749\uA74B\uA74D\uA74F\uA751\uA753\uA755\uA757\uA759\uA75B\uA75D\uA75F\uA761\uA763\uA765\uA767\uA769\uA76B\uA76D\uA76F\uA771-\uA778\uA77A\uA77C\uA77F\uA781\uA783\uA785\uA787\uA78C\uA78E\uA791\uA793-\uA795\uA797\uA799\uA79B\uA79D\uA79F\uA7A1\uA7A3\uA7A5\uA7A7\uA7A9\uA7AF\uA7B5\uA7B7\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E01\u1E03\u1E05\u1E07\u1E09\u1E0B\u1E0D\u1E0F\u1E11\u1E13\u1E15\u1E17\u1E19\u1E1B\u1E1D\u1E1F\u1E21\u1E23\u1E25\u1E27\u1E29\u1E2B\u1E2D\u1E2F\u1E31\u1E33\u1E35\u1E37\u1E39\u1E3B\u1E3D\u1E3F\u1E41\u1E43\u1E45\u1E47\u1E49\u1E4B\u1E4D\u1E4F\u1E51\u1E53\u1E55\u1E57\u1E59\u1E5B\u1E5D\u1E5F\u1E61\u1E63\u1E65\u1E67\u1E69\u1E6B\u1E6D\u1E6F\u1E71\u1E73\u1E75\u1E77\u1E79\u1E7B\u1E7D\u1E7F\u1E81\u1E83\u1E85\u1E87\u1E89\u1E8B\u1E8D\u1E8F\u1E91\u1E93\u1E95-\u1E9D\u1E9F\u1EA1\u1EA3\u1EA5\u1EA7\u1EA9\u1EAB\u1EAD\u1EAF\u1EB1\u1EB3\u1EB5\u1EB7\u1EB9\u1EBB\u1EBD\u1EBF\u1EC1\u1EC3\u1EC5\u1EC7\u1EC9\u1ECB\u1ECD\u1ECF\u1ED1\u1ED3\u1ED5\u1ED7\u1ED9\u1EDB\u1EDD\u1EDF\u1EE1\u1EE3\u1EE5\u1EE7\u1EE9\u1EEB\u1EED\u1EEF\u1EF1\u1EF3\u1EF5\u1EF7\u1EF9\u1EFB\u1EFD\u1EFFёа-яәөүҗңһα-ωάέίόώήύа-щюяіїєґѓѕјљњќѐѝ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F\'"”“`‘´’‚,„»«「」『』（）〔〕【】《》〈〉〈〉⟦⟧])\.(?=[A-Z\uFF21-\uFF3A\u00C0-\u00D6\u00D8-\u00DE\u0100\u0102\u0104\u0106\u0108\u010A\u010C\u010E\u0110\u0112\u0114\u0116\u0118\u011A\u011C\u011E\u0120\u0122\u0124\u0126\u0128\u012A\u012C\u012E\u0130\u0132\u0134\u0136\u0139\u013B\u013D\u013F\u0141\u0143\u0145\u0147\u014A\u014C\u014E\u0150\u0152\u0154\u0156\u0158\u015A\u015C\u015E\u0160\u0162\u0164\u0166\u0168\u016A\u016C\u016E\u0170\u0172\u0174\u0176\u0178\u0179\u017B\u017D\u0181\u0182\u0184\u0186\u0187\u0189-\u018B\u018E-\u0191\u0193\u0194\u0196-\u0198\u019C\u019D\u019F\u01A0\u01A2\u01A4\u01A6\u01A7\u01A9\u01AC\u01AE\u01AF\u01B1-\u01B3\u01B5\u01B7\u01B8\u01BC\u01C4\u01C7\u01CA\u01CD\u01CF\u01D1\u01D3\u01D5\u01D7\u01D9\u01DB\u01DE\u01E0\u01E2\u01E4\u01E6\u01E8\u01EA\u01EC\u01EE\u01F1\u01F4\u01F6-\u01F8\u01FA\u01FC\u01FE\u0200\u0202\u0204\u0206\u0208\u020A\u020C\u020E\u0210\u0212\u0214\u0216\u0218\u021A\u021C\u021E\u0220\u0222\u0224\u0226\u0228\u022A\u022C\u022E\u0230\u0232\u023A\u023B\u023D\u023E\u0241\u0243-\u0246\u0248\u024A\u024C\u024E\u2C60\u2C62-\u2C64\u2C67\u2C69\u2C6B\u2C6D-\u2C70\u2C72\u2C75\u2C7E\u2C7F\uA722\uA724\uA726\uA728\uA72A\uA72C\uA72E\uA732\uA734\uA736\uA738\uA73A\uA73C\uA73E\uA740\uA742\uA744\uA746\uA748\uA74A\uA74C\uA74E\uA750\uA752\uA754\uA756\uA758\uA75A\uA75C\uA75E\uA760\uA762\uA764\uA766\uA768\uA76A\uA76C\uA76E\uA779\uA77B\uA77D\uA77E\uA780\uA782\uA784\uA786\uA78B\uA78D\uA790\uA792\uA796\uA798\uA79A\uA79C\uA79E\uA7A0\uA7A2\uA7A4\uA7A6\uA7A8\uA7AA-\uA7AE\uA7B0-\uA7B4\uA7B6\uA7B8\u1E00\u1E02\u1E04\u1E06\u1E08\u1E0A\u1E0C\u1E0E\u1E10\u1E12\u1E14\u1E16\u1E18\u1E1A\u1E1C\u1E1E\u1E20\u1E22\u1E24\u1E26\u1E28\u1E2A\u1E2C\u1E2E\u1E30\u1E32\u1E34\u1E36\u1E38\u1E3A\u1E3C\u1E3E\u1E40\u1E42\u1E44\u1E46\u1E48\u1E4A\u1E4C\u1E4E\u1E50\u1E52\u1E54\u1E56\u1E58\u1E5A\u1E5C\u1E5E\u1E60\u1E62\u1E64\u1E66\u1E68\u1E6A\u1E6C\u1E6E\u1E70\u1E72\u1E74\u1E76\u1E78\u1E7A\u1E7C\u1E7E\u1E80\u1E82\u1E84\u1E86\u1E88\u1E8A\u1E8C\u1E8E\u1E90\u1E92\u1E94\u1E9E\u1EA0\u1EA2\u1EA4\u1EA6\u1EA8\u1EAA\u1EAC\u1EAE\u1EB0\u1EB2\u1EB4\u1EB6\u1EB8\u1EBA\u1EBC\u1EBE\u1EC0\u1EC2\u1EC4\u1EC6\u1EC8\u1ECA\u1ECC\u1ECE\u1ED0\u1ED2\u1ED4\u1ED6\u1ED8\u1EDA\u1EDC\u1EDE\u1EE0\u1EE2\u1EE4\u1EE6\u1EE8\u1EEA\u1EEC\u1EEE\u1EF0\u1EF2\u1EF4\u1EF6\u1EF8\u1EFA\u1EFC\u1EFEЁА-ЯӘӨҮҖҢҺΑ-ΩΆΈΊΌΏΉΎА-ЩЮЯІЇЄҐЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F\'"”“`‘´’‚,„»«「」『』（）〔〕【】《》〈〉〈〉⟦⟧])|(?<=[A-Za-z\uFF21-\uFF3A\uFF41-\uFF5A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u017F\u0180-\u01BF\u01C4-\u024F\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFFёа-яЁА-ЯәөүҗңһӘӨҮҖҢҺα-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎа-щюяіїєґА-ЩЮЯІЇЄҐѓѕјљњќѐѝЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F]),(?=[A-Za-z\uFF21-\uFF3A\uFF41-\uFF5A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u017F\u0180-\u01BF\u01C4-\u024F\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFFёа-яЁА-ЯәөүҗңһӘӨҮҖҢҺα-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎа-щюяіїєґА-ЩЮЯІЇЄҐѓѕјљњќѐѝЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])|(?<=[A-Za-z\uFF21-\uFF3A\uFF41-\uFF5A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u017F\u0180-\u01BF\u01C4-\u024F\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFFёа-яЁА-ЯәөүҗңһӘӨҮҖҢҺα-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎа-щюяіїєґА-ЩЮЯІЇЄҐѓѕјљњќѐѝЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])(?:-|–|—|--|---|——|~)(?=[A-Za-z\uFF21-\uFF3A\uFF41-\uFF5A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u017F\u0180-\u01BF\u01C4-\u024F\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFFёа-яЁА-ЯәөүҗңһӘӨҮҖҢҺα-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎа-щюяіїєґА-ЩЮЯІЇЄҐѓѕјљњќѐѝЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])|(?<=[A-Za-z\uFF21-\uFF3A\uFF41-\uFF5A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u017F\u0180-\u01BF\u01C4-\u024F\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFFёа-яЁА-ЯәөүҗңһӘӨҮҖҢҺα-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎа-щюяіїєґА-ЩЮЯІЇЄҐѓѕјљњќѐѝЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F0-9])[:<>=/](?=[A-Za-z\uFF21-\uFF3A\uFF41-\uFF5A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u017F\u0180-\u01BF\u01C4-\u024F\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFFёа-яЁА-ЯәөүҗңһӘӨҮҖҢҺα-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎа-щюяіїєґА-ЩЮЯІЇЄҐѓѕјљњќѐѝЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])�token_match��url_match�
2
+ ��A�
3
+ � ��A� �'��A�'�''��A�''�(*_*)��A�(*_*)�(-8��A�(-8�(-:��A�(-:�(-;��A�(-;�(-_-)��A�(-_-)�(._.)��A�(._.)�(:��A�(:�(;��A�(;�(=��A�(=�(>_<)��A�(>_<)�(^_^)��A�(^_^)�(o:��A�(o:�(¬_¬)��A�(¬_¬)�(ಠ_ಠ)��A�(ಠ_ಠ)�(╯°□°）╯︵┻━┻��A�(╯°□°）╯︵┻━┻�)-:��A�)-:�):��A�):�-_-��A�-_-�-__-��A�-__-�._.��A�._.�0.0��A�0.0�0.o��A�0.o�0_0��A�0_0�0_o��A�0_o�8)��A�8)�8-)��A�8-)�8-D��A�8-D�8D��A�8D�:'(��A�:'(�:')��A�:')�:'-(��A�:'-(�:'-)��A�:'-)�:(��A�:(�:((��A�:((�:(((��A�:(((�:()��A�:()�:)��A�:)�:))��A�:))�:)))��A�:)))�:*��A�:*�:-(��A�:-(�:-((��A�:-((�:-(((��A�:-(((�:-)��A�:-)�:-))��A�:-))�:-)))��A�:-)))�:-*��A�:-*�:-/��A�:-/�:-0��A�:-0�:-3��A�:-3�:->��A�:->�:-D��A�:-D�:-O��A�:-O�:-P��A�:-P�:-X��A�:-X�:-]��A�:-]�:-o��A�:-o�:-p��A�:-p�:-x��A�:-x�:-|��A�:-|�:-}��A�:-}�:/��A�:/�:0��A�:0�:1��A�:1�:3��A�:3�:>��A�:>�:D��A�:D�:O��A�:O�:P��A�:P�:X��A�:X�:]��A�:]�:o��A�:o�:o)��A�:o)�:p��A�:p�:x��A�:x�:|��A�:|�:}��A�:}�;)��A�;)�;-)��A�;-)�;-D��A�;-D�;D��A�;D�;_;��A�;_;�<.<��A�<.<�</3��A�</3�<3��A�<3�<33��A�<33�<333��A�<333�<space>��A�<space>�=(��A�=(�=)��A�=)�=/��A�=/�=3��A�=3�=D��A�=D�=[��A�=[�=]��A�=]�=|��A�=|�>.<��A�>.<�>.>��A�>.>�>:(��A�>:(�>:o��A�>:o�><(((*>��A�><(((*>�@_@��A�@_@�C++��A�C++�O.O��A�O.O�O.o��A�O.o�O_O��A�O_O�O_o��A�O_o�V.V��A�V.V�V_V��A�V_V�XD��A�XD�XDD��A�XDD�[-:��A�[-:�[:��A�[:�[=��A�[=�\")��A�\")�\n��A�\n�\t��A�\t�]=��A�]=�^_^��A�^_^�^__^��A�^__^�^___^��A�^___^�a.��A�a.�b.��A�b.�c.��A�c.�d.��A�d.�e.��A�e.�f.��A�f.�g.��A�g.�h.��A�h.�i.��A�i.�j.��A�j.�k.��A�k.�l.��A�l.�m.��A�m.�n.��A�n.�o.��A�o.�o.0��A�o.0�o.O��A�o.O�o.o��A�o.o�o_0��A�o_0�o_O��A�o_O�o_o��A�o_o�p.��A�p.�q.��A�q.�r.��A�r.�s.��A�s.�t.��A�t.�u.��A�u.�v.��A�v.�v.v��A�v.v�v_v��A�v_v�w.��A�w.�x.��A�x.�xD��A�xD�xDD��A�xDD�y.��A�y.�z.��A�z.� ��A� C� �¯\(ツ)/¯��A�¯\(ツ)/¯�°C.��A�°�A�C�A�.�°F.��A�°�A�F�A�.�°K.��A�°�A�K�A�.�°c.��A�°�A�c�A�.�°f.��A�°�A�f�A�.�°k.��A�°�A�k�A�.�ä.��A�ä.�ö.��A�ö.�ü.��A�ü.�ಠ_ಠ��A�ಠ_ಠ�ಠ︵ಠ��A�ಠ︵ಠ�—��A�—�faster_heuristics�
trainable_lemmatizer/cfg ADDED
@@ -0,0 +1,215 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "labels":[
3
+ 0,
4
+ 2,
5
+ 4,
6
+ 6,
7
+ 8,
8
+ 9,
9
+ 11,
10
+ 14,
11
+ 16,
12
+ 17,
13
+ 19,
14
+ 22,
15
+ 23,
16
+ 24,
17
+ 26,
18
+ 27,
19
+ 28,
20
+ 30,
21
+ 31,
22
+ 35,
23
+ 37,
24
+ 40,
25
+ 42,
26
+ 44,
27
+ 47,
28
+ 48,
29
+ 50,
30
+ 52,
31
+ 54,
32
+ 56,
33
+ 57,
34
+ 58,
35
+ 59,
36
+ 61,
37
+ 64,
38
+ 67,
39
+ 71,
40
+ 74,
41
+ 78,
42
+ 49,
43
+ 79,
44
+ 82,
45
+ 83,
46
+ 85,
47
+ 13,
48
+ 87,
49
+ 90,
50
+ 91,
51
+ 93,
52
+ 95,
53
+ 96,
54
+ 100,
55
+ 101,
56
+ 104,
57
+ 107,
58
+ 109,
59
+ 111,
60
+ 113,
61
+ 115,
62
+ 117,
63
+ 118,
64
+ 120,
65
+ 51,
66
+ 121,
67
+ 124,
68
+ 128,
69
+ 131,
70
+ 133,
71
+ 136,
72
+ 138,
73
+ 139,
74
+ 143,
75
+ 145,
76
+ 147,
77
+ 149,
78
+ 151,
79
+ 153,
80
+ 154,
81
+ 156,
82
+ 158,
83
+ 161,
84
+ 162,
85
+ 165,
86
+ 166,
87
+ 169,
88
+ 172,
89
+ 174,
90
+ 175,
91
+ 180,
92
+ 181,
93
+ 183,
94
+ 184,
95
+ 186,
96
+ 187,
97
+ 188,
98
+ 189,
99
+ 191,
100
+ 195,
101
+ 196,
102
+ 198,
103
+ 200,
104
+ 202,
105
+ 204,
106
+ 206,
107
+ 208,
108
+ 210,
109
+ 212,
110
+ 214,
111
+ 216,
112
+ 217,
113
+ 218,
114
+ 108,
115
+ 221,
116
+ 223,
117
+ 226,
118
+ 228,
119
+ 150,
120
+ 230,
121
+ 233,
122
+ 234,
123
+ 235,
124
+ 239,
125
+ 173,
126
+ 241,
127
+ 244,
128
+ 245,
129
+ 247,
130
+ 250,
131
+ 252,
132
+ 255,
133
+ 63,
134
+ 256,
135
+ 259,
136
+ 262,
137
+ 264,
138
+ 266,
139
+ 270,
140
+ 272,
141
+ 273,
142
+ 274,
143
+ 275,
144
+ 276,
145
+ 278,
146
+ 280,
147
+ 281,
148
+ 283,
149
+ 286,
150
+ 290,
151
+ 293,
152
+ 295,
153
+ 297,
154
+ 298,
155
+ 299,
156
+ 301,
157
+ 303,
158
+ 306,
159
+ 309,
160
+ 311,
161
+ 312,
162
+ 314,
163
+ 317,
164
+ 321,
165
+ 323,
166
+ 324,
167
+ 326,
168
+ 328,
169
+ 330,
170
+ 332,
171
+ 333,
172
+ 334,
173
+ 336,
174
+ 337,
175
+ 338,
176
+ 340,
177
+ 343,
178
+ 345,
179
+ 347,
180
+ 349,
181
+ 352,
182
+ 354,
183
+ 358,
184
+ 361,
185
+ 362,
186
+ 363,
187
+ 364,
188
+ 368,
189
+ 370,
190
+ 372,
191
+ 373,
192
+ 374,
193
+ 376,
194
+ 377,
195
+ 379,
196
+ 381,
197
+ 382,
198
+ 386,
199
+ 388,
200
+ 73,
201
+ 389,
202
+ 391,
203
+ 393,
204
+ 395,
205
+ 397,
206
+ 399,
207
+ 401,
208
+ 402,
209
+ 404,
210
+ 407,
211
+ 408,
212
+ 409,
213
+ 322
214
+ ]
215
+ }
trainable_lemmatizer/model ADDED
Binary file (650 kB). View file
 
trainable_lemmatizer/trees ADDED
Binary file (18.4 kB). View file
 
transformer/cfg ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "max_batch_items":4096
3
+ }
transformer/model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46be0d5dc0f52a847b51237dbd608c4f024e7aa8c1121b2fc2226d7300e42fc4
3
+ size 508514878
vocab/key2row ADDED
@@ -0,0 +1 @@
 
 
1
+ �
vocab/lookups.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76be8b528d0075f7aae98d6fa57a6d3c83ae480a8469e668d7b0af968995ac71
3
+ size 1
vocab/strings.json ADDED
The diff for this file is too large to render. See raw diff
 
vocab/vectors ADDED
Binary file (128 Bytes). View file
 
vocab/vectors.cfg ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "mode":"default"
3
+ }
xx_cop_sigtyp_trf-any-py3-none-any.whl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:983a92be90346f7cabf9e0de7c23430428868dc220752d2222d639957983ef84
3
+ size 470089744