KoichiYasuoka commited on
Commit
4b9af83
·
1 Parent(s): 131f4b9

initial release

Browse files
Files changed (8) hide show
  1. README.md +30 -0
  2. config.json +2265 -0
  3. maker.py +138 -0
  4. pytorch_model.bin +3 -0
  5. special_tokens_map.json +1249 -0
  6. tokenizer.json +0 -0
  7. tokenizer_config.json +0 -0
  8. ud.py +142 -0
README.md ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - "th"
4
+ tags:
5
+ - "thai"
6
+ - "pos"
7
+ - "dependency-parsing"
8
+ base_model: goldfish-models/tha_thai_1000mb
9
+ datasets:
10
+ - "universal_dependencies"
11
+ license: "apache-2.0"
12
+ pipeline_tag: "token-classification"
13
+ widget:
14
+ - text: "หลายหัวดีกว่าหัวเดียว"
15
+ ---
16
+
17
+ # goldfish-gpt2-thai-ud-causal
18
+
19
+ ## Model Description
20
+
21
+ This is a GPT-2 model pre-trained for POS-tagging and dependency-parsing, derived from [tha_thai_1000mb](https://huggingface.co/goldfish-models/tha_thai_1000mb)refined for [Thai Universal Dependency Treebank](https://github.com/nlp-chula/TUD).
22
+
23
+ ## How to Use
24
+
25
+ ```py
26
+ from transformers import pipeline
27
+ nlp=pipeline("universal-dependencies","KoichiYasuoka/goldfish-gpt2-thai-ud-causal",trust_remote_code=True)
28
+ print(nlp("หลายหัวดีกว่าหัวเดียว"))
29
+ ```
30
+
config.json ADDED
@@ -0,0 +1,2265 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu",
3
+ "architectures": [
4
+ "GPT2ForTokenClassification"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50000,
8
+ "custom_pipelines": {
9
+ "upos": {
10
+ "impl": "ud.BellmanFordTokenClassificationPipeline",
11
+ "pt": "AutoModelForTokenClassification"
12
+ },
13
+ "universal-dependencies": {
14
+ "impl": "ud.UniversalDependenciesCausalPipeline",
15
+ "pt": "AutoModelForTokenClassification"
16
+ }
17
+ },
18
+ "embd_pdrop": 0.1,
19
+ "eos_token_id": 50001,
20
+ "id2label": {
21
+ "0": "ADP",
22
+ "1": "ADP|Foreign=Yes",
23
+ "2": "ADP|Foreign=Yes|l-case",
24
+ "3": "ADP|NounType=Class",
25
+ "4": "ADP|NounType=Class|l-case",
26
+ "5": "ADP|Prefix=Yes",
27
+ "6": "ADP|Prefix=Yes|l-case",
28
+ "7": "ADP|Prefix=Yes|l-mark",
29
+ "8": "ADP|l-acl",
30
+ "9": "ADP|l-advcl",
31
+ "10": "ADP|l-advmod",
32
+ "11": "ADP|l-case",
33
+ "12": "ADP|l-cc",
34
+ "13": "ADP|l-dep",
35
+ "14": "ADP|l-fixed",
36
+ "15": "ADP|l-flat",
37
+ "16": "ADP|l-mark",
38
+ "17": "ADP|l-nmod",
39
+ "18": "ADP|l-nsubj",
40
+ "19": "ADP|l-obl",
41
+ "20": "ADP|l-orphan",
42
+ "21": "ADP|r-acl",
43
+ "22": "ADP|r-advmod",
44
+ "23": "ADP|r-case",
45
+ "24": "ADP|r-compound",
46
+ "25": "ADP|r-conj",
47
+ "26": "ADP|r-fixed",
48
+ "27": "ADP|r-flat",
49
+ "28": "ADP|r-mark",
50
+ "29": "ADP|r-obl",
51
+ "30": "ADP|r-orphan",
52
+ "31": "ADP|root",
53
+ "32": "ADV",
54
+ "33": "ADV|Foreign=Yes",
55
+ "34": "ADV|Foreign=Yes|l-advmod",
56
+ "35": "ADV|Foreign=Yes|r-advmod",
57
+ "36": "ADV|NumType=Mult",
58
+ "37": "ADV|NumType=Mult|r-advmod",
59
+ "38": "ADV|PartType=Adv",
60
+ "39": "ADV|PartType=Adv|l-advmod",
61
+ "40": "ADV|PartType=Adv|l-mark",
62
+ "41": "ADV|PartType=Adv|r-advmod",
63
+ "42": "ADV|PartType=Enp",
64
+ "43": "ADV|PartType=Enp|l-advmod",
65
+ "44": "ADV|PartType=Enp|r-advmod",
66
+ "45": "ADV|PartType=Int",
67
+ "46": "ADV|PartType=Int|r-advmod",
68
+ "47": "ADV|PartType=Int|r-fixed",
69
+ "48": "ADV|Prefix=Yes",
70
+ "49": "ADV|Prefix=Yes|l-advmod",
71
+ "50": "ADV|Prefix=Yes|l-mark",
72
+ "51": "ADV|Prefix=Yes|r-advmod",
73
+ "52": "ADV|l-acl",
74
+ "53": "ADV|l-advcl",
75
+ "54": "ADV|l-advmod",
76
+ "55": "ADV|l-aux",
77
+ "56": "ADV|l-case",
78
+ "57": "ADV|l-compound",
79
+ "58": "ADV|l-dep",
80
+ "59": "ADV|l-det",
81
+ "60": "ADV|l-discourse",
82
+ "61": "ADV|l-fixed",
83
+ "62": "ADV|l-mark",
84
+ "63": "ADV|l-orphan",
85
+ "64": "ADV|r-acl",
86
+ "65": "ADV|r-advcl",
87
+ "66": "ADV|r-advmod",
88
+ "67": "ADV|r-aux",
89
+ "68": "ADV|r-ccomp",
90
+ "69": "ADV|r-compound",
91
+ "70": "ADV|r-conj",
92
+ "71": "ADV|r-det",
93
+ "72": "ADV|r-fixed",
94
+ "73": "ADV|r-flat",
95
+ "74": "ADV|r-mark",
96
+ "75": "ADV|r-nmod",
97
+ "76": "ADV|r-obj",
98
+ "77": "ADV|r-orphan",
99
+ "78": "ADV|r-xcomp",
100
+ "79": "ADV|root",
101
+ "80": "AUX",
102
+ "81": "AUX|Foreign=Yes",
103
+ "82": "AUX|Foreign=Yes|l-aux",
104
+ "83": "AUX|NounType=Class",
105
+ "84": "AUX|NounType=Class|r-appos",
106
+ "85": "AUX|Prefix=Yes",
107
+ "86": "AUX|Prefix=Yes|l-aux",
108
+ "87": "AUX|Prefix=Yes|r-aux",
109
+ "88": "AUX|VerbType=Cop",
110
+ "89": "AUX|VerbType=Cop|l-acl",
111
+ "90": "AUX|VerbType=Cop|l-advcl",
112
+ "91": "AUX|VerbType=Cop|l-aux",
113
+ "92": "AUX|VerbType=Cop|l-cop",
114
+ "93": "AUX|VerbType=Cop|r-acl",
115
+ "94": "AUX|VerbType=Cop|r-advcl",
116
+ "95": "AUX|VerbType=Cop|r-aux",
117
+ "96": "AUX|VerbType=Cop|r-conj",
118
+ "97": "AUX|VerbType=Cop|r-mark",
119
+ "98": "AUX|VerbType=Cop|root",
120
+ "99": "AUX|l-advmod",
121
+ "100": "AUX|l-aux",
122
+ "101": "AUX|l-aux:pass",
123
+ "102": "AUX|l-cop",
124
+ "103": "AUX|l-mark",
125
+ "104": "AUX|r-acl",
126
+ "105": "AUX|r-advmod",
127
+ "106": "AUX|r-aux",
128
+ "107": "AUX|r-ccomp",
129
+ "108": "AUX|r-clf",
130
+ "109": "AUX|r-compound",
131
+ "110": "AUX|r-conj",
132
+ "111": "AUX|r-fixed",
133
+ "112": "AUX|r-mark",
134
+ "113": "AUX|root",
135
+ "114": "B-ADP",
136
+ "115": "B-ADP|Foreign=Yes",
137
+ "116": "B-ADP|NounType=Class",
138
+ "117": "B-ADP|Prefix=Yes",
139
+ "118": "B-ADV",
140
+ "119": "B-ADV|Foreign=Yes",
141
+ "120": "B-ADV|NumType=Mult",
142
+ "121": "B-ADV|PartType=Adv",
143
+ "122": "B-ADV|PartType=Enp",
144
+ "123": "B-ADV|PartType=Int",
145
+ "124": "B-ADV|Prefix=Yes",
146
+ "125": "B-AUX",
147
+ "126": "B-AUX|Foreign=Yes",
148
+ "127": "B-AUX|NounType=Class",
149
+ "128": "B-AUX|Prefix=Yes",
150
+ "129": "B-AUX|VerbType=Cop",
151
+ "130": "B-CCONJ",
152
+ "131": "B-CCONJ|Foreign=Yes",
153
+ "132": "B-CCONJ|PronType=Prs",
154
+ "133": "B-DET",
155
+ "134": "B-DET|NumType=Mult",
156
+ "135": "B-DET|PartType=Emp",
157
+ "136": "B-DET|PartType=Int",
158
+ "137": "B-DET|PronType=Int",
159
+ "138": "B-INTJ",
160
+ "139": "B-NOUN",
161
+ "140": "B-NOUN|Abbr=Yes",
162
+ "141": "B-NOUN|Abbr=Yes|Foreign=Yes",
163
+ "142": "B-NOUN|Abbr=Yes|Prefix=Yes",
164
+ "143": "B-NOUN|Foreign=Yes",
165
+ "144": "B-NOUN|Foreign=Yes|NounType=Class",
166
+ "145": "B-NOUN|Foreign=Yes|Prefix=Yes",
167
+ "146": "B-NOUN|NameType=Com",
168
+ "147": "B-NOUN|NameType=Geo",
169
+ "148": "B-NOUN|NameType=Nat",
170
+ "149": "B-NOUN|NameType=Oth",
171
+ "150": "B-NOUN|NameType=Pro",
172
+ "151": "B-NOUN|NameType=Prs",
173
+ "152": "B-NOUN|NounType=Class",
174
+ "153": "B-NOUN|NounType=Class|Prefix=Yes",
175
+ "154": "B-NOUN|NumType=Mult",
176
+ "155": "B-NOUN|PartType=Enp",
177
+ "156": "B-NOUN|PartType=Int",
178
+ "157": "B-NOUN|PartType=Res",
179
+ "158": "B-NOUN|Prefix=Yes",
180
+ "159": "B-NUM",
181
+ "160": "B-NUM|Abbr=Yes",
182
+ "161": "B-NUM|Foreign=Yes",
183
+ "162": "B-NUM|NumType=Mult",
184
+ "163": "B-NUM|Prefix=Yes",
185
+ "164": "B-PART",
186
+ "165": "B-PART|Aspect=Perf",
187
+ "166": "B-PART|Aspect=Prog",
188
+ "167": "B-PART|NameType=Oth",
189
+ "168": "B-PART|NounType=Class|PartType=Emp",
190
+ "169": "B-PART|NounType=Class|PartType=Emp|Prefix=Yes",
191
+ "170": "B-PART|NounType=Class|Prefix=Yes",
192
+ "171": "B-PART|NumType=Mult|PartType=Emp",
193
+ "172": "B-PART|PartType=Adj",
194
+ "173": "B-PART|PartType=Adv",
195
+ "174": "B-PART|PartType=Emp",
196
+ "175": "B-PART|PartType=Emp|Prefix=Yes",
197
+ "176": "B-PART|PartType=Enp",
198
+ "177": "B-PART|PartType=Int",
199
+ "178": "B-PART|PartType=Neg",
200
+ "179": "B-PART|PartType=Res",
201
+ "180": "B-PART|Polarity=Neg",
202
+ "181": "B-PART|Prefix=Yes",
203
+ "182": "B-PART|PronType=Int",
204
+ "183": "B-PRON",
205
+ "184": "B-PRON|NounType=Class",
206
+ "185": "B-PRON|Person=1",
207
+ "186": "B-PRON|Person=2",
208
+ "187": "B-PRON|Person=3",
209
+ "188": "B-PRON|PronType=Int",
210
+ "189": "B-PRON|PronType=Prs",
211
+ "190": "B-PRON|PronType=Rcp",
212
+ "191": "B-PROPN",
213
+ "192": "B-PROPN|Abbr=Yes",
214
+ "193": "B-PROPN|Abbr=Yes|Foreign=Yes|NameType=Oth",
215
+ "194": "B-PROPN|Abbr=Yes|NameType=Com",
216
+ "195": "B-PROPN|Foreign=Yes",
217
+ "196": "B-PROPN|Foreign=Yes|NameType=Com",
218
+ "197": "B-PROPN|Foreign=Yes|NameType=Geo",
219
+ "198": "B-PROPN|Foreign=Yes|NameType=Giv",
220
+ "199": "B-PROPN|Foreign=Yes|NameType=Oth",
221
+ "200": "B-PROPN|Foreign=Yes|NameType=Prs",
222
+ "201": "B-PROPN|Foreign=Yes|NameType=Sur",
223
+ "202": "B-PROPN|NameType=Com",
224
+ "203": "B-PROPN|NameType=Geo",
225
+ "204": "B-PROPN|NameType=Giv",
226
+ "205": "B-PROPN|NameType=Nat",
227
+ "206": "B-PROPN|NameType=Oth",
228
+ "207": "B-PROPN|NameType=Pro",
229
+ "208": "B-PROPN|NameType=Prs",
230
+ "209": "B-PROPN|NameType=Sur",
231
+ "210": "B-PROPN|NounType=Class",
232
+ "211": "B-PROPN|Prefix=Yes",
233
+ "212": "B-PUNCT",
234
+ "213": "B-PUNCT|NounType=Class",
235
+ "214": "B-SCONJ",
236
+ "215": "B-SCONJ|NumType=Mult",
237
+ "216": "B-SCONJ|Prefix=Yes",
238
+ "217": "B-SCONJ|VerbType=Cop",
239
+ "218": "B-SYM",
240
+ "219": "B-VERB",
241
+ "220": "B-VERB|Abbr=Yes",
242
+ "221": "B-VERB|Foreign=Yes",
243
+ "222": "B-VERB|Mood=Imp",
244
+ "223": "B-VERB|NounType=Class",
245
+ "224": "B-VERB|PartType=Adj",
246
+ "225": "B-VERB|Prefix=Yes",
247
+ "226": "B-VERB|VerbType=Cop",
248
+ "227": "B-VERB|Voice=Pass",
249
+ "228": "B-X",
250
+ "229": "CCONJ",
251
+ "230": "CCONJ|Foreign=Yes",
252
+ "231": "CCONJ|Foreign=Yes|l-cc",
253
+ "232": "CCONJ|PronType=Prs",
254
+ "233": "CCONJ|PronType=Prs|l-cc",
255
+ "234": "CCONJ|l-advmod",
256
+ "235": "CCONJ|l-case",
257
+ "236": "CCONJ|l-cc",
258
+ "237": "CCONJ|l-conj",
259
+ "238": "CCONJ|l-discourse",
260
+ "239": "CCONJ|l-fixed",
261
+ "240": "CCONJ|l-flat",
262
+ "241": "CCONJ|l-mark",
263
+ "242": "CCONJ|l-nsubj",
264
+ "243": "CCONJ|l-obj",
265
+ "244": "CCONJ|l-orphan",
266
+ "245": "CCONJ|r-cc",
267
+ "246": "CCONJ|r-compound",
268
+ "247": "CCONJ|r-fixed",
269
+ "248": "CCONJ|r-mark",
270
+ "249": "CCONJ|r-obl",
271
+ "250": "CCONJ|root",
272
+ "251": "DET",
273
+ "252": "DET|NumType=Mult",
274
+ "253": "DET|NumType=Mult|l-det",
275
+ "254": "DET|PartType=Emp",
276
+ "255": "DET|PartType=Emp|r-det",
277
+ "256": "DET|PartType=Int",
278
+ "257": "DET|PartType=Int|r-det",
279
+ "258": "DET|PronType=Int",
280
+ "259": "DET|l-advmod",
281
+ "260": "DET|l-case",
282
+ "261": "DET|l-compound",
283
+ "262": "DET|l-det",
284
+ "263": "DET|l-discourse",
285
+ "264": "DET|l-nsubj",
286
+ "265": "DET|l-obl",
287
+ "266": "DET|l-orphan",
288
+ "267": "DET|r-advmod",
289
+ "268": "DET|r-compound",
290
+ "269": "DET|r-conj",
291
+ "270": "DET|r-dep",
292
+ "271": "DET|r-det",
293
+ "272": "DET|r-fixed",
294
+ "273": "DET|r-flat",
295
+ "274": "DET|r-list",
296
+ "275": "DET|r-nmod",
297
+ "276": "DET|r-nummod",
298
+ "277": "DET|r-obj",
299
+ "278": "DET|r-obl",
300
+ "279": "DET|r-orphan",
301
+ "280": "DET|root",
302
+ "281": "I-ADP",
303
+ "282": "I-ADP|Foreign=Yes",
304
+ "283": "I-ADP|NounType=Class",
305
+ "284": "I-ADP|Prefix=Yes",
306
+ "285": "I-ADV",
307
+ "286": "I-ADV|Foreign=Yes",
308
+ "287": "I-ADV|NumType=Mult",
309
+ "288": "I-ADV|PartType=Adv",
310
+ "289": "I-ADV|PartType=Enp",
311
+ "290": "I-ADV|PartType=Int",
312
+ "291": "I-ADV|Prefix=Yes",
313
+ "292": "I-AUX",
314
+ "293": "I-AUX|Foreign=Yes",
315
+ "294": "I-AUX|NounType=Class",
316
+ "295": "I-AUX|Prefix=Yes",
317
+ "296": "I-AUX|VerbType=Cop",
318
+ "297": "I-CCONJ",
319
+ "298": "I-CCONJ|Foreign=Yes",
320
+ "299": "I-CCONJ|PronType=Prs",
321
+ "300": "I-DET",
322
+ "301": "I-DET|NumType=Mult",
323
+ "302": "I-DET|PartType=Emp",
324
+ "303": "I-DET|PartType=Int",
325
+ "304": "I-DET|PronType=Int",
326
+ "305": "I-INTJ",
327
+ "306": "I-NOUN",
328
+ "307": "I-NOUN|Abbr=Yes",
329
+ "308": "I-NOUN|Abbr=Yes|Foreign=Yes",
330
+ "309": "I-NOUN|Abbr=Yes|Prefix=Yes",
331
+ "310": "I-NOUN|Foreign=Yes",
332
+ "311": "I-NOUN|Foreign=Yes|NounType=Class",
333
+ "312": "I-NOUN|Foreign=Yes|Prefix=Yes",
334
+ "313": "I-NOUN|NameType=Com",
335
+ "314": "I-NOUN|NameType=Geo",
336
+ "315": "I-NOUN|NameType=Nat",
337
+ "316": "I-NOUN|NameType=Oth",
338
+ "317": "I-NOUN|NameType=Pro",
339
+ "318": "I-NOUN|NameType=Prs",
340
+ "319": "I-NOUN|NounType=Class",
341
+ "320": "I-NOUN|NounType=Class|Prefix=Yes",
342
+ "321": "I-NOUN|NumType=Mult",
343
+ "322": "I-NOUN|PartType=Enp",
344
+ "323": "I-NOUN|PartType=Int",
345
+ "324": "I-NOUN|PartType=Res",
346
+ "325": "I-NOUN|Prefix=Yes",
347
+ "326": "I-NUM",
348
+ "327": "I-NUM|Abbr=Yes",
349
+ "328": "I-NUM|Foreign=Yes",
350
+ "329": "I-NUM|NumType=Mult",
351
+ "330": "I-NUM|Prefix=Yes",
352
+ "331": "I-PART",
353
+ "332": "I-PART|Aspect=Perf",
354
+ "333": "I-PART|Aspect=Prog",
355
+ "334": "I-PART|NameType=Oth",
356
+ "335": "I-PART|NounType=Class|PartType=Emp",
357
+ "336": "I-PART|NounType=Class|PartType=Emp|Prefix=Yes",
358
+ "337": "I-PART|NounType=Class|Prefix=Yes",
359
+ "338": "I-PART|NumType=Mult|PartType=Emp",
360
+ "339": "I-PART|PartType=Adj",
361
+ "340": "I-PART|PartType=Adv",
362
+ "341": "I-PART|PartType=Emp",
363
+ "342": "I-PART|PartType=Emp|Prefix=Yes",
364
+ "343": "I-PART|PartType=Enp",
365
+ "344": "I-PART|PartType=Int",
366
+ "345": "I-PART|PartType=Neg",
367
+ "346": "I-PART|PartType=Res",
368
+ "347": "I-PART|Polarity=Neg",
369
+ "348": "I-PART|Prefix=Yes",
370
+ "349": "I-PART|PronType=Int",
371
+ "350": "I-PRON",
372
+ "351": "I-PRON|NounType=Class",
373
+ "352": "I-PRON|Person=1",
374
+ "353": "I-PRON|Person=2",
375
+ "354": "I-PRON|Person=3",
376
+ "355": "I-PRON|PronType=Int",
377
+ "356": "I-PRON|PronType=Prs",
378
+ "357": "I-PRON|PronType=Rcp",
379
+ "358": "I-PROPN",
380
+ "359": "I-PROPN|Abbr=Yes",
381
+ "360": "I-PROPN|Abbr=Yes|Foreign=Yes|NameType=Oth",
382
+ "361": "I-PROPN|Abbr=Yes|NameType=Com",
383
+ "362": "I-PROPN|Foreign=Yes",
384
+ "363": "I-PROPN|Foreign=Yes|NameType=Com",
385
+ "364": "I-PROPN|Foreign=Yes|NameType=Geo",
386
+ "365": "I-PROPN|Foreign=Yes|NameType=Giv",
387
+ "366": "I-PROPN|Foreign=Yes|NameType=Oth",
388
+ "367": "I-PROPN|Foreign=Yes|NameType=Prs",
389
+ "368": "I-PROPN|Foreign=Yes|NameType=Sur",
390
+ "369": "I-PROPN|NameType=Com",
391
+ "370": "I-PROPN|NameType=Geo",
392
+ "371": "I-PROPN|NameType=Giv",
393
+ "372": "I-PROPN|NameType=Nat",
394
+ "373": "I-PROPN|NameType=Oth",
395
+ "374": "I-PROPN|NameType=Pro",
396
+ "375": "I-PROPN|NameType=Prs",
397
+ "376": "I-PROPN|NameType=Sur",
398
+ "377": "I-PROPN|NounType=Class",
399
+ "378": "I-PROPN|Prefix=Yes",
400
+ "379": "I-PUNCT",
401
+ "380": "I-PUNCT|NounType=Class",
402
+ "381": "I-SCONJ",
403
+ "382": "I-SCONJ|NumType=Mult",
404
+ "383": "I-SCONJ|Prefix=Yes",
405
+ "384": "I-SCONJ|VerbType=Cop",
406
+ "385": "I-SYM",
407
+ "386": "I-VERB",
408
+ "387": "I-VERB|Abbr=Yes",
409
+ "388": "I-VERB|Foreign=Yes",
410
+ "389": "I-VERB|Mood=Imp",
411
+ "390": "I-VERB|NounType=Class",
412
+ "391": "I-VERB|PartType=Adj",
413
+ "392": "I-VERB|Prefix=Yes",
414
+ "393": "I-VERB|VerbType=Cop",
415
+ "394": "I-VERB|Voice=Pass",
416
+ "395": "I-X",
417
+ "396": "INTJ",
418
+ "397": "INTJ|l-nsubj",
419
+ "398": "INTJ|r-acl",
420
+ "399": "INTJ|root",
421
+ "400": "NOUN",
422
+ "401": "NOUN|Abbr=Yes",
423
+ "402": "NOUN|Abbr=Yes|Foreign=Yes",
424
+ "403": "NOUN|Abbr=Yes|Foreign=Yes|r-nmod",
425
+ "404": "NOUN|Abbr=Yes|Prefix=Yes",
426
+ "405": "NOUN|Abbr=Yes|Prefix=Yes|l-flat",
427
+ "406": "NOUN|Abbr=Yes|l-flat",
428
+ "407": "NOUN|Abbr=Yes|l-nmod",
429
+ "408": "NOUN|Abbr=Yes|l-nsubj",
430
+ "409": "NOUN|Abbr=Yes|l-obl",
431
+ "410": "NOUN|Abbr=Yes|r-acl",
432
+ "411": "NOUN|Abbr=Yes|r-appos",
433
+ "412": "NOUN|Abbr=Yes|r-clf",
434
+ "413": "NOUN|Abbr=Yes|r-conj",
435
+ "414": "NOUN|Abbr=Yes|r-fixed",
436
+ "415": "NOUN|Abbr=Yes|r-flat",
437
+ "416": "NOUN|Abbr=Yes|r-nmod",
438
+ "417": "NOUN|Abbr=Yes|r-obj",
439
+ "418": "NOUN|Abbr=Yes|r-obl",
440
+ "419": "NOUN|Foreign=Yes",
441
+ "420": "NOUN|Foreign=Yes|NounType=Class",
442
+ "421": "NOUN|Foreign=Yes|NounType=Class|r-clf",
443
+ "422": "NOUN|Foreign=Yes|NounType=Class|r-obj",
444
+ "423": "NOUN|Foreign=Yes|Prefix=Yes",
445
+ "424": "NOUN|Foreign=Yes|Prefix=Yes|l-flat",
446
+ "425": "NOUN|Foreign=Yes|Prefix=Yes|r-appos",
447
+ "426": "NOUN|Foreign=Yes|l-dislocated",
448
+ "427": "NOUN|Foreign=Yes|l-flat",
449
+ "428": "NOUN|Foreign=Yes|l-nmod",
450
+ "429": "NOUN|Foreign=Yes|l-nsubj",
451
+ "430": "NOUN|Foreign=Yes|l-obl",
452
+ "431": "NOUN|Foreign=Yes|r-acl",
453
+ "432": "NOUN|Foreign=Yes|r-advcl",
454
+ "433": "NOUN|Foreign=Yes|r-advmod",
455
+ "434": "NOUN|Foreign=Yes|r-appos",
456
+ "435": "NOUN|Foreign=Yes|r-ccomp",
457
+ "436": "NOUN|Foreign=Yes|r-clf",
458
+ "437": "NOUN|Foreign=Yes|r-compound",
459
+ "438": "NOUN|Foreign=Yes|r-conj",
460
+ "439": "NOUN|Foreign=Yes|r-flat",
461
+ "440": "NOUN|Foreign=Yes|r-iobj",
462
+ "441": "NOUN|Foreign=Yes|r-list",
463
+ "442": "NOUN|Foreign=Yes|r-nmod",
464
+ "443": "NOUN|Foreign=Yes|r-obj",
465
+ "444": "NOUN|Foreign=Yes|r-obl",
466
+ "445": "NOUN|Foreign=Yes|r-xcomp",
467
+ "446": "NOUN|Foreign=Yes|root",
468
+ "447": "NOUN|NameType=Com",
469
+ "448": "NOUN|NameType=Com|r-nmod",
470
+ "449": "NOUN|NameType=Geo",
471
+ "450": "NOUN|NameType=Geo|l-nsubj",
472
+ "451": "NOUN|NameType=Geo|r-nmod",
473
+ "452": "NOUN|NameType=Geo|r-obj",
474
+ "453": "NOUN|NameType=Nat",
475
+ "454": "NOUN|NameType=Nat|r-nmod",
476
+ "455": "NOUN|NameType=Oth",
477
+ "456": "NOUN|NameType=Oth|l-nsubj",
478
+ "457": "NOUN|NameType=Oth|r-conj",
479
+ "458": "NOUN|NameType=Oth|r-flat",
480
+ "459": "NOUN|NameType=Oth|r-nmod",
481
+ "460": "NOUN|NameType=Pro",
482
+ "461": "NOUN|NameType=Pro|r-nmod",
483
+ "462": "NOUN|NameType=Prs",
484
+ "463": "NOUN|NameType=Prs|l-nsubj",
485
+ "464": "NOUN|NameType=Prs|r-nmod",
486
+ "465": "NOUN|NounType=Class",
487
+ "466": "NOUN|NounType=Class|Prefix=Yes",
488
+ "467": "NOUN|NounType=Class|Prefix=Yes|l-advcl",
489
+ "468": "NOUN|NounType=Class|Prefix=Yes|l-advmod",
490
+ "469": "NOUN|NounType=Class|Prefix=Yes|l-mark",
491
+ "470": "NOUN|NounType=Class|Prefix=Yes|l-nmod",
492
+ "471": "NOUN|NounType=Class|Prefix=Yes|l-nsubj",
493
+ "472": "NOUN|NounType=Class|Prefix=Yes|r-advcl",
494
+ "473": "NOUN|NounType=Class|Prefix=Yes|r-clf",
495
+ "474": "NOUN|NounType=Class|Prefix=Yes|r-nmod",
496
+ "475": "NOUN|NounType=Class|Prefix=Yes|r-obj",
497
+ "476": "NOUN|NounType=Class|l-advcl",
498
+ "477": "NOUN|NounType=Class|l-advmod",
499
+ "478": "NOUN|NounType=Class|l-clf",
500
+ "479": "NOUN|NounType=Class|l-dislocated",
501
+ "480": "NOUN|NounType=Class|l-nmod",
502
+ "481": "NOUN|NounType=Class|l-nsubj",
503
+ "482": "NOUN|NounType=Class|l-obj",
504
+ "483": "NOUN|NounType=Class|l-obl",
505
+ "484": "NOUN|NounType=Class|r-acl",
506
+ "485": "NOUN|NounType=Class|r-advcl",
507
+ "486": "NOUN|NounType=Class|r-advmod",
508
+ "487": "NOUN|NounType=Class|r-appos",
509
+ "488": "NOUN|NounType=Class|r-cc",
510
+ "489": "NOUN|NounType=Class|r-ccomp",
511
+ "490": "NOUN|NounType=Class|r-clf",
512
+ "491": "NOUN|NounType=Class|r-compound",
513
+ "492": "NOUN|NounType=Class|r-conj",
514
+ "493": "NOUN|NounType=Class|r-dislocated",
515
+ "494": "NOUN|NounType=Class|r-fixed",
516
+ "495": "NOUN|NounType=Class|r-flat",
517
+ "496": "NOUN|NounType=Class|r-iobj",
518
+ "497": "NOUN|NounType=Class|r-list",
519
+ "498": "NOUN|NounType=Class|r-nmod",
520
+ "499": "NOUN|NounType=Class|r-nummod",
521
+ "500": "NOUN|NounType=Class|r-obj",
522
+ "501": "NOUN|NounType=Class|r-obl",
523
+ "502": "NOUN|NounType=Class|r-orphan",
524
+ "503": "NOUN|NounType=Class|r-xcomp",
525
+ "504": "NOUN|NounType=Class|root",
526
+ "505": "NOUN|NumType=Mult",
527
+ "506": "NOUN|NumType=Mult|r-advcl",
528
+ "507": "NOUN|NumType=Mult|r-nmod",
529
+ "508": "NOUN|NumType=Mult|r-obj",
530
+ "509": "NOUN|PartType=Enp",
531
+ "510": "NOUN|PartType=Enp|r-obj",
532
+ "511": "NOUN|PartType=Enp|r-obl",
533
+ "512": "NOUN|PartType=Int",
534
+ "513": "NOUN|PartType=Int|r-obj",
535
+ "514": "NOUN|PartType=Res",
536
+ "515": "NOUN|PartType=Res|r-nmod",
537
+ "516": "NOUN|PartType=Res|r-obj",
538
+ "517": "NOUN|Prefix=Yes",
539
+ "518": "NOUN|Prefix=Yes|l-acl",
540
+ "519": "NOUN|Prefix=Yes|l-advcl",
541
+ "520": "NOUN|Prefix=Yes|l-clf",
542
+ "521": "NOUN|Prefix=Yes|l-csubj",
543
+ "522": "NOUN|Prefix=Yes|l-dislocated",
544
+ "523": "NOUN|Prefix=Yes|l-flat",
545
+ "524": "NOUN|Prefix=Yes|l-nmod",
546
+ "525": "NOUN|Prefix=Yes|l-nsubj",
547
+ "526": "NOUN|Prefix=Yes|l-obj",
548
+ "527": "NOUN|Prefix=Yes|l-obl",
549
+ "528": "NOUN|Prefix=Yes|r-acl",
550
+ "529": "NOUN|Prefix=Yes|r-advcl",
551
+ "530": "NOUN|Prefix=Yes|r-advmod",
552
+ "531": "NOUN|Prefix=Yes|r-appos",
553
+ "532": "NOUN|Prefix=Yes|r-case",
554
+ "533": "NOUN|Prefix=Yes|r-cc",
555
+ "534": "NOUN|Prefix=Yes|r-ccomp",
556
+ "535": "NOUN|Prefix=Yes|r-clf",
557
+ "536": "NOUN|Prefix=Yes|r-compound",
558
+ "537": "NOUN|Prefix=Yes|r-conj",
559
+ "538": "NOUN|Prefix=Yes|r-dislocated",
560
+ "539": "NOUN|Prefix=Yes|r-fixed",
561
+ "540": "NOUN|Prefix=Yes|r-flat",
562
+ "541": "NOUN|Prefix=Yes|r-iobj",
563
+ "542": "NOUN|Prefix=Yes|r-list",
564
+ "543": "NOUN|Prefix=Yes|r-nmod",
565
+ "544": "NOUN|Prefix=Yes|r-nummod",
566
+ "545": "NOUN|Prefix=Yes|r-obj",
567
+ "546": "NOUN|Prefix=Yes|r-obl",
568
+ "547": "NOUN|Prefix=Yes|r-orphan",
569
+ "548": "NOUN|Prefix=Yes|r-xcomp",
570
+ "549": "NOUN|Prefix=Yes|root",
571
+ "550": "NOUN|l-acl",
572
+ "551": "NOUN|l-advcl",
573
+ "552": "NOUN|l-advmod",
574
+ "553": "NOUN|l-aux",
575
+ "554": "NOUN|l-case",
576
+ "555": "NOUN|l-ccomp",
577
+ "556": "NOUN|l-compound",
578
+ "557": "NOUN|l-csubj",
579
+ "558": "NOUN|l-discourse",
580
+ "559": "NOUN|l-dislocated",
581
+ "560": "NOUN|l-expl",
582
+ "561": "NOUN|l-flat",
583
+ "562": "NOUN|l-iobj",
584
+ "563": "NOUN|l-mark",
585
+ "564": "NOUN|l-nmod",
586
+ "565": "NOUN|l-nsubj",
587
+ "566": "NOUN|l-nummod",
588
+ "567": "NOUN|l-obj",
589
+ "568": "NOUN|l-obl",
590
+ "569": "NOUN|l-orphan",
591
+ "570": "NOUN|l-vocative",
592
+ "571": "NOUN|r-acl",
593
+ "572": "NOUN|r-advcl",
594
+ "573": "NOUN|r-advmod",
595
+ "574": "NOUN|r-appos",
596
+ "575": "NOUN|r-case",
597
+ "576": "NOUN|r-cc",
598
+ "577": "NOUN|r-ccomp",
599
+ "578": "NOUN|r-clf",
600
+ "579": "NOUN|r-compound",
601
+ "580": "NOUN|r-conj",
602
+ "581": "NOUN|r-cop",
603
+ "582": "NOUN|r-discourse",
604
+ "583": "NOUN|r-dislocated",
605
+ "584": "NOUN|r-fixed",
606
+ "585": "NOUN|r-flat",
607
+ "586": "NOUN|r-flat:name",
608
+ "587": "NOUN|r-iobj",
609
+ "588": "NOUN|r-list",
610
+ "589": "NOUN|r-mark",
611
+ "590": "NOUN|r-nmod",
612
+ "591": "NOUN|r-nmod:poss",
613
+ "592": "NOUN|r-nsubj",
614
+ "593": "NOUN|r-nummod",
615
+ "594": "NOUN|r-obj",
616
+ "595": "NOUN|r-obl",
617
+ "596": "NOUN|r-obl:tmod",
618
+ "597": "NOUN|r-orphan",
619
+ "598": "NOUN|r-parataxis",
620
+ "599": "NOUN|r-xcomp",
621
+ "600": "NOUN|root",
622
+ "601": "NUM",
623
+ "602": "NUM|Abbr=Yes",
624
+ "603": "NUM|Abbr=Yes|r-flat",
625
+ "604": "NUM|Abbr=Yes|r-nummod",
626
+ "605": "NUM|Abbr=Yes|r-obj",
627
+ "606": "NUM|Foreign=Yes",
628
+ "607": "NUM|Foreign=Yes|r-clf",
629
+ "608": "NUM|NumType=Mult",
630
+ "609": "NUM|NumType=Mult|l-advmod",
631
+ "610": "NUM|NumType=Mult|l-nummod",
632
+ "611": "NUM|NumType=Mult|r-advmod",
633
+ "612": "NUM|Prefix=Yes",
634
+ "613": "NUM|Prefix=Yes|l-nummod",
635
+ "614": "NUM|l-advcl",
636
+ "615": "NUM|l-advmod",
637
+ "616": "NUM|l-case",
638
+ "617": "NUM|l-clf",
639
+ "618": "NUM|l-dep",
640
+ "619": "NUM|l-flat",
641
+ "620": "NUM|l-nmod",
642
+ "621": "NUM|l-nsubj",
643
+ "622": "NUM|l-nummod",
644
+ "623": "NUM|l-obl",
645
+ "624": "NUM|r-acl",
646
+ "625": "NUM|r-advmod",
647
+ "626": "NUM|r-appos",
648
+ "627": "NUM|r-ccomp",
649
+ "628": "NUM|r-clf",
650
+ "629": "NUM|r-compound",
651
+ "630": "NUM|r-conj",
652
+ "631": "NUM|r-det",
653
+ "632": "NUM|r-fixed",
654
+ "633": "NUM|r-flat",
655
+ "634": "NUM|r-flat:name",
656
+ "635": "NUM|r-iobj",
657
+ "636": "NUM|r-nmod",
658
+ "637": "NUM|r-nummod",
659
+ "638": "NUM|r-obj",
660
+ "639": "NUM|r-obl",
661
+ "640": "NUM|r-obl:tmod",
662
+ "641": "NUM|root",
663
+ "642": "PART",
664
+ "643": "PART|Aspect=Perf",
665
+ "644": "PART|Aspect=Prog",
666
+ "645": "PART|NameType=Oth",
667
+ "646": "PART|NameType=Oth|l-advmod",
668
+ "647": "PART|NounType=Class|PartType=Emp",
669
+ "648": "PART|NounType=Class|PartType=Emp|Prefix=Yes",
670
+ "649": "PART|NounType=Class|PartType=Emp|Prefix=Yes|l-mark",
671
+ "650": "PART|NounType=Class|PartType=Emp|l-mark",
672
+ "651": "PART|NounType=Class|Prefix=Yes",
673
+ "652": "PART|NounType=Class|Prefix=Yes|l-mark",
674
+ "653": "PART|NumType=Mult|PartType=Emp",
675
+ "654": "PART|NumType=Mult|PartType=Emp|l-mark",
676
+ "655": "PART|PartType=Adj",
677
+ "656": "PART|PartType=Adj|l-mark",
678
+ "657": "PART|PartType=Adj|l-orphan",
679
+ "658": "PART|PartType=Adj|r-acl",
680
+ "659": "PART|PartType=Adj|r-compound",
681
+ "660": "PART|PartType=Adj|r-nmod",
682
+ "661": "PART|PartType=Adv",
683
+ "662": "PART|PartType=Adv|l-advmod",
684
+ "663": "PART|PartType=Adv|l-mark",
685
+ "664": "PART|PartType=Adv|r-advmod",
686
+ "665": "PART|PartType=Emp",
687
+ "666": "PART|PartType=Emp|Prefix=Yes",
688
+ "667": "PART|PartType=Emp|Prefix=Yes|l-advmod",
689
+ "668": "PART|PartType=Emp|Prefix=Yes|l-aux",
690
+ "669": "PART|PartType=Emp|Prefix=Yes|l-mark",
691
+ "670": "PART|PartType=Emp|l-advmod",
692
+ "671": "PART|PartType=Emp|l-case",
693
+ "672": "PART|PartType=Emp|l-discourse",
694
+ "673": "PART|PartType=Emp|l-mark",
695
+ "674": "PART|PartType=Emp|r-acl",
696
+ "675": "PART|PartType=Emp|r-advmod",
697
+ "676": "PART|PartType=Emp|r-aux",
698
+ "677": "PART|PartType=Emp|r-compound",
699
+ "678": "PART|PartType=Emp|r-det",
700
+ "679": "PART|PartType=Emp|r-fixed",
701
+ "680": "PART|PartType=Emp|r-mark",
702
+ "681": "PART|PartType=Emp|r-nmod",
703
+ "682": "PART|PartType=Enp",
704
+ "683": "PART|PartType=Enp|l-discourse",
705
+ "684": "PART|PartType=Enp|r-acl",
706
+ "685": "PART|PartType=Enp|r-advmod",
707
+ "686": "PART|PartType=Enp|r-compound",
708
+ "687": "PART|PartType=Enp|r-dep",
709
+ "688": "PART|PartType=Enp|r-det",
710
+ "689": "PART|PartType=Enp|r-discourse",
711
+ "690": "PART|PartType=Enp|r-fixed",
712
+ "691": "PART|PartType=Enp|r-obl",
713
+ "692": "PART|PartType=Int",
714
+ "693": "PART|PartType=Int|l-advmod",
715
+ "694": "PART|PartType=Int|l-mark",
716
+ "695": "PART|PartType=Int|r-acl",
717
+ "696": "PART|PartType=Int|r-advmod",
718
+ "697": "PART|PartType=Int|r-dep",
719
+ "698": "PART|PartType=Int|r-discourse",
720
+ "699": "PART|PartType=Int|r-nmod",
721
+ "700": "PART|PartType=Int|r-obj",
722
+ "701": "PART|PartType=Int|r-obl",
723
+ "702": "PART|PartType=Neg",
724
+ "703": "PART|PartType=Neg|l-advcl",
725
+ "704": "PART|PartType=Neg|l-advmod",
726
+ "705": "PART|PartType=Neg|l-aux",
727
+ "706": "PART|PartType=Neg|l-mark",
728
+ "707": "PART|PartType=Neg|r-acl",
729
+ "708": "PART|PartType=Neg|r-advmod",
730
+ "709": "PART|PartType=Neg|r-fixed",
731
+ "710": "PART|PartType=Res",
732
+ "711": "PART|PartType=Res|r-advmod",
733
+ "712": "PART|PartType=Res|r-discourse",
734
+ "713": "PART|PartType=Res|r-fixed",
735
+ "714": "PART|Polarity=Neg",
736
+ "715": "PART|Polarity=Neg|l-advmod",
737
+ "716": "PART|Prefix=Yes",
738
+ "717": "PART|Prefix=Yes|l-advmod",
739
+ "718": "PART|Prefix=Yes|l-aux",
740
+ "719": "PART|Prefix=Yes|l-mark",
741
+ "720": "PART|Prefix=Yes|r-acl",
742
+ "721": "PART|Prefix=Yes|r-nmod",
743
+ "722": "PART|PronType=Int",
744
+ "723": "PART|PronType=Int|r-acl",
745
+ "724": "PART|PronType=Int|r-advmod",
746
+ "725": "PART|PronType=Int|r-obj",
747
+ "726": "PART|PronType=Int|root",
748
+ "727": "PART|l-advmod",
749
+ "728": "PART|l-cc",
750
+ "729": "PART|l-discourse",
751
+ "730": "PART|l-mark",
752
+ "731": "PART|l-nsubj",
753
+ "732": "PART|r-acl",
754
+ "733": "PART|r-advmod",
755
+ "734": "PART|r-ccomp",
756
+ "735": "PART|r-clf",
757
+ "736": "PART|r-compound",
758
+ "737": "PART|r-conj",
759
+ "738": "PART|r-discourse",
760
+ "739": "PART|r-fixed",
761
+ "740": "PART|r-mark",
762
+ "741": "PART|r-nmod",
763
+ "742": "PART|r-nmod:poss",
764
+ "743": "PART|r-obj",
765
+ "744": "PART|r-obl",
766
+ "745": "PART|root",
767
+ "746": "PRON",
768
+ "747": "PRON|NounType=Class",
769
+ "748": "PRON|NounType=Class|r-clf",
770
+ "749": "PRON|Person=1",
771
+ "750": "PRON|Person=1|l-nsubj",
772
+ "751": "PRON|Person=1|r-compound",
773
+ "752": "PRON|Person=1|r-nmod:poss",
774
+ "753": "PRON|Person=2",
775
+ "754": "PRON|Person=2|r-compound",
776
+ "755": "PRON|Person=3",
777
+ "756": "PRON|Person=3|l-nsubj",
778
+ "757": "PRON|Person=3|r-appos",
779
+ "758": "PRON|Person=3|r-compound",
780
+ "759": "PRON|Person=3|r-nmod:poss",
781
+ "760": "PRON|Person=3|r-obj",
782
+ "761": "PRON|PronType=Int",
783
+ "762": "PRON|PronType=Int|l-nsubj",
784
+ "763": "PRON|PronType=Int|r-obj",
785
+ "764": "PRON|PronType=Int|r-obl",
786
+ "765": "PRON|PronType=Prs",
787
+ "766": "PRON|PronType=Prs|l-advmod",
788
+ "767": "PRON|PronType=Prs|l-expl",
789
+ "768": "PRON|PronType=Prs|l-nsubj",
790
+ "769": "PRON|PronType=Prs|l-obj",
791
+ "770": "PRON|PronType=Prs|l-obl",
792
+ "771": "PRON|PronType=Prs|r-advcl",
793
+ "772": "PRON|PronType=Prs|r-advmod",
794
+ "773": "PRON|PronType=Prs|r-ccomp",
795
+ "774": "PRON|PronType=Prs|r-clf",
796
+ "775": "PRON|PronType=Prs|r-conj",
797
+ "776": "PRON|PronType=Prs|r-nmod",
798
+ "777": "PRON|PronType=Prs|r-nsubj",
799
+ "778": "PRON|PronType=Prs|r-obj",
800
+ "779": "PRON|PronType=Prs|r-obl",
801
+ "780": "PRON|PronType=Prs|root",
802
+ "781": "PRON|PronType=Rcp",
803
+ "782": "PRON|PronType=Rcp|r-advmod",
804
+ "783": "PRON|PronType=Rcp|r-iobj",
805
+ "784": "PRON|PronType=Rcp|r-nmod",
806
+ "785": "PRON|PronType=Rcp|r-obj",
807
+ "786": "PRON|PronType=Rcp|r-obl",
808
+ "787": "PRON|l-advcl",
809
+ "788": "PRON|l-advmod",
810
+ "789": "PRON|l-compound",
811
+ "790": "PRON|l-csubj",
812
+ "791": "PRON|l-dislocated",
813
+ "792": "PRON|l-expl",
814
+ "793": "PRON|l-iobj",
815
+ "794": "PRON|l-mark",
816
+ "795": "PRON|l-nsubj",
817
+ "796": "PRON|l-obj",
818
+ "797": "PRON|l-obl",
819
+ "798": "PRON|r-acl",
820
+ "799": "PRON|r-advmod",
821
+ "800": "PRON|r-appos",
822
+ "801": "PRON|r-ccomp",
823
+ "802": "PRON|r-compound",
824
+ "803": "PRON|r-conj",
825
+ "804": "PRON|r-det",
826
+ "805": "PRON|r-discourse",
827
+ "806": "PRON|r-fixed",
828
+ "807": "PRON|r-flat",
829
+ "808": "PRON|r-iobj",
830
+ "809": "PRON|r-nmod",
831
+ "810": "PRON|r-nsubj",
832
+ "811": "PRON|r-obj",
833
+ "812": "PRON|r-obl",
834
+ "813": "PRON|r-xcomp",
835
+ "814": "PROPN",
836
+ "815": "PROPN|Abbr=Yes",
837
+ "816": "PROPN|Abbr=Yes|Foreign=Yes|NameType=Oth",
838
+ "817": "PROPN|Abbr=Yes|Foreign=Yes|NameType=Oth|r-obj",
839
+ "818": "PROPN|Abbr=Yes|NameType=Com",
840
+ "819": "PROPN|Abbr=Yes|NameType=Com|r-advmod",
841
+ "820": "PROPN|Abbr=Yes|NameType=Com|r-nmod",
842
+ "821": "PROPN|Abbr=Yes|l-nmod",
843
+ "822": "PROPN|Abbr=Yes|l-nsubj",
844
+ "823": "PROPN|Abbr=Yes|r-nmod",
845
+ "824": "PROPN|Foreign=Yes",
846
+ "825": "PROPN|Foreign=Yes|NameType=Com",
847
+ "826": "PROPN|Foreign=Yes|NameType=Com|l-nsubj",
848
+ "827": "PROPN|Foreign=Yes|NameType=Com|r-list",
849
+ "828": "PROPN|Foreign=Yes|NameType=Com|r-nmod",
850
+ "829": "PROPN|Foreign=Yes|NameType=Com|r-obl",
851
+ "830": "PROPN|Foreign=Yes|NameType=Geo",
852
+ "831": "PROPN|Foreign=Yes|NameType=Geo|r-obj",
853
+ "832": "PROPN|Foreign=Yes|NameType=Geo|r-obl",
854
+ "833": "PROPN|Foreign=Yes|NameType=Giv",
855
+ "834": "PROPN|Foreign=Yes|NameType=Giv|l-nsubj",
856
+ "835": "PROPN|Foreign=Yes|NameType=Oth",
857
+ "836": "PROPN|Foreign=Yes|NameType=Oth|r-conj",
858
+ "837": "PROPN|Foreign=Yes|NameType=Oth|r-flat",
859
+ "838": "PROPN|Foreign=Yes|NameType=Oth|r-nmod",
860
+ "839": "PROPN|Foreign=Yes|NameType=Prs",
861
+ "840": "PROPN|Foreign=Yes|NameType=Prs|l-flat",
862
+ "841": "PROPN|Foreign=Yes|NameType=Prs|l-nsubj",
863
+ "842": "PROPN|Foreign=Yes|NameType=Prs|r-conj",
864
+ "843": "PROPN|Foreign=Yes|NameType=Prs|r-flat",
865
+ "844": "PROPN|Foreign=Yes|NameType=Prs|r-nmod",
866
+ "845": "PROPN|Foreign=Yes|NameType=Prs|r-obj",
867
+ "846": "PROPN|Foreign=Yes|NameType=Prs|r-obl",
868
+ "847": "PROPN|Foreign=Yes|NameType=Sur",
869
+ "848": "PROPN|Foreign=Yes|NameType=Sur|r-flat",
870
+ "849": "PROPN|Foreign=Yes|l-flat",
871
+ "850": "PROPN|Foreign=Yes|l-nmod",
872
+ "851": "PROPN|Foreign=Yes|l-nsubj",
873
+ "852": "PROPN|Foreign=Yes|l-obl",
874
+ "853": "PROPN|Foreign=Yes|r-appos",
875
+ "854": "PROPN|Foreign=Yes|r-ccomp",
876
+ "855": "PROPN|Foreign=Yes|r-compound",
877
+ "856": "PROPN|Foreign=Yes|r-conj",
878
+ "857": "PROPN|Foreign=Yes|r-flat",
879
+ "858": "PROPN|Foreign=Yes|r-iobj",
880
+ "859": "PROPN|Foreign=Yes|r-list",
881
+ "860": "PROPN|Foreign=Yes|r-nmod",
882
+ "861": "PROPN|Foreign=Yes|r-nsubj",
883
+ "862": "PROPN|Foreign=Yes|r-obj",
884
+ "863": "PROPN|Foreign=Yes|r-obl",
885
+ "864": "PROPN|Foreign=Yes|root",
886
+ "865": "PROPN|NameType=Com",
887
+ "866": "PROPN|NameType=Com|l-nsubj",
888
+ "867": "PROPN|NameType=Com|l-obl",
889
+ "868": "PROPN|NameType=Com|r-appos",
890
+ "869": "PROPN|NameType=Com|r-conj",
891
+ "870": "PROPN|NameType=Com|r-flat",
892
+ "871": "PROPN|NameType=Com|r-list",
893
+ "872": "PROPN|NameType=Com|r-nmod",
894
+ "873": "PROPN|NameType=Com|r-nsubj",
895
+ "874": "PROPN|NameType=Com|r-obj",
896
+ "875": "PROPN|NameType=Com|r-obl",
897
+ "876": "PROPN|NameType=Geo",
898
+ "877": "PROPN|NameType=Geo|l-nsubj",
899
+ "878": "PROPN|NameType=Geo|l-obl",
900
+ "879": "PROPN|NameType=Geo|r-compound",
901
+ "880": "PROPN|NameType=Geo|r-conj",
902
+ "881": "PROPN|NameType=Geo|r-flat",
903
+ "882": "PROPN|NameType=Geo|r-list",
904
+ "883": "PROPN|NameType=Geo|r-nmod",
905
+ "884": "PROPN|NameType=Geo|r-nsubj",
906
+ "885": "PROPN|NameType=Geo|r-nummod",
907
+ "886": "PROPN|NameType=Geo|r-obj",
908
+ "887": "PROPN|NameType=Geo|r-obl",
909
+ "888": "PROPN|NameType=Geo|root",
910
+ "889": "PROPN|NameType=Giv",
911
+ "890": "PROPN|NameType=Giv|l-dislocated",
912
+ "891": "PROPN|NameType=Giv|l-nsubj",
913
+ "892": "PROPN|NameType=Giv|l-obl",
914
+ "893": "PROPN|NameType=Giv|r-acl",
915
+ "894": "PROPN|NameType=Giv|r-appos",
916
+ "895": "PROPN|NameType=Giv|r-ccomp",
917
+ "896": "PROPN|NameType=Giv|r-conj",
918
+ "897": "PROPN|NameType=Giv|r-flat",
919
+ "898": "PROPN|NameType=Giv|r-list",
920
+ "899": "PROPN|NameType=Giv|r-nmod",
921
+ "900": "PROPN|NameType=Giv|r-nsubj",
922
+ "901": "PROPN|NameType=Giv|r-obj",
923
+ "902": "PROPN|NameType=Giv|r-obl",
924
+ "903": "PROPN|NameType=Giv|root",
925
+ "904": "PROPN|NameType=Nat",
926
+ "905": "PROPN|NameType=Nat|l-csubj",
927
+ "906": "PROPN|NameType=Nat|l-nsubj",
928
+ "907": "PROPN|NameType=Nat|l-obl",
929
+ "908": "PROPN|NameType=Nat|r-acl",
930
+ "909": "PROPN|NameType=Nat|r-appos",
931
+ "910": "PROPN|NameType=Nat|r-compound",
932
+ "911": "PROPN|NameType=Nat|r-conj",
933
+ "912": "PROPN|NameType=Nat|r-flat",
934
+ "913": "PROPN|NameType=Nat|r-list",
935
+ "914": "PROPN|NameType=Nat|r-nmod",
936
+ "915": "PROPN|NameType=Nat|r-nummod",
937
+ "916": "PROPN|NameType=Nat|r-obj",
938
+ "917": "PROPN|NameType=Nat|r-obl",
939
+ "918": "PROPN|NameType=Oth",
940
+ "919": "PROPN|NameType=Oth|l-dislocated",
941
+ "920": "PROPN|NameType=Oth|l-nsubj",
942
+ "921": "PROPN|NameType=Oth|r-acl",
943
+ "922": "PROPN|NameType=Oth|r-appos",
944
+ "923": "PROPN|NameType=Oth|r-compound",
945
+ "924": "PROPN|NameType=Oth|r-conj",
946
+ "925": "PROPN|NameType=Oth|r-flat",
947
+ "926": "PROPN|NameType=Oth|r-nmod",
948
+ "927": "PROPN|NameType=Oth|r-obj",
949
+ "928": "PROPN|NameType=Oth|r-obl",
950
+ "929": "PROPN|NameType=Oth|root",
951
+ "930": "PROPN|NameType=Pro",
952
+ "931": "PROPN|NameType=Pro|l-nsubj",
953
+ "932": "PROPN|NameType=Pro|l-obl",
954
+ "933": "PROPN|NameType=Pro|r-advcl",
955
+ "934": "PROPN|NameType=Pro|r-flat",
956
+ "935": "PROPN|NameType=Pro|r-nmod",
957
+ "936": "PROPN|NameType=Pro|r-obj",
958
+ "937": "PROPN|NameType=Prs",
959
+ "938": "PROPN|NameType=Prs|l-dislocated",
960
+ "939": "PROPN|NameType=Prs|l-nsubj",
961
+ "940": "PROPN|NameType=Prs|l-obl",
962
+ "941": "PROPN|NameType=Prs|l-vocative",
963
+ "942": "PROPN|NameType=Prs|r-conj",
964
+ "943": "PROPN|NameType=Prs|r-discourse",
965
+ "944": "PROPN|NameType=Prs|r-flat",
966
+ "945": "PROPN|NameType=Prs|r-list",
967
+ "946": "PROPN|NameType=Prs|r-nmod",
968
+ "947": "PROPN|NameType=Prs|r-obj",
969
+ "948": "PROPN|NameType=Prs|r-obl",
970
+ "949": "PROPN|NameType=Prs|r-vocative",
971
+ "950": "PROPN|NameType=Sur",
972
+ "951": "PROPN|NameType=Sur|l-nsubj",
973
+ "952": "PROPN|NameType=Sur|r-flat",
974
+ "953": "PROPN|NameType=Sur|r-nmod",
975
+ "954": "PROPN|NounType=Class",
976
+ "955": "PROPN|NounType=Class|r-clf",
977
+ "956": "PROPN|Prefix=Yes",
978
+ "957": "PROPN|Prefix=Yes|l-nsubj",
979
+ "958": "PROPN|Prefix=Yes|r-nmod",
980
+ "959": "PROPN|l-advmod",
981
+ "960": "PROPN|l-aux",
982
+ "961": "PROPN|l-nsubj",
983
+ "962": "PROPN|l-obl",
984
+ "963": "PROPN|r-acl",
985
+ "964": "PROPN|r-advmod",
986
+ "965": "PROPN|r-appos",
987
+ "966": "PROPN|r-clf",
988
+ "967": "PROPN|r-compound",
989
+ "968": "PROPN|r-conj",
990
+ "969": "PROPN|r-fixed",
991
+ "970": "PROPN|r-flat",
992
+ "971": "PROPN|r-flat:name",
993
+ "972": "PROPN|r-iobj",
994
+ "973": "PROPN|r-list",
995
+ "974": "PROPN|r-nmod",
996
+ "975": "PROPN|r-nmod:poss",
997
+ "976": "PROPN|r-obj",
998
+ "977": "PROPN|r-obl",
999
+ "978": "PROPN|root",
1000
+ "979": "PUNCT",
1001
+ "980": "PUNCT|NounType=Class",
1002
+ "981": "PUNCT|NounType=Class|r-punct",
1003
+ "982": "PUNCT|l-advmod",
1004
+ "983": "PUNCT|l-dep",
1005
+ "984": "PUNCT|l-punct",
1006
+ "985": "PUNCT|r-advmod",
1007
+ "986": "PUNCT|r-clf",
1008
+ "987": "PUNCT|r-dep",
1009
+ "988": "PUNCT|r-punct",
1010
+ "989": "PUNCT|root",
1011
+ "990": "SCONJ",
1012
+ "991": "SCONJ|NumType=Mult",
1013
+ "992": "SCONJ|NumType=Mult|l-mark",
1014
+ "993": "SCONJ|Prefix=Yes",
1015
+ "994": "SCONJ|Prefix=Yes|l-cc",
1016
+ "995": "SCONJ|Prefix=Yes|l-mark",
1017
+ "996": "SCONJ|VerbType=Cop",
1018
+ "997": "SCONJ|VerbType=Cop|l-mark",
1019
+ "998": "SCONJ|l-advmod",
1020
+ "999": "SCONJ|l-case",
1021
+ "1000": "SCONJ|l-cc",
1022
+ "1001": "SCONJ|l-discourse",
1023
+ "1002": "SCONJ|l-mark",
1024
+ "1003": "SCONJ|l-nsubj",
1025
+ "1004": "SCONJ|l-orphan",
1026
+ "1005": "SCONJ|r-advcl",
1027
+ "1006": "SCONJ|r-compound",
1028
+ "1007": "SCONJ|r-fixed",
1029
+ "1008": "SCONJ|r-flat",
1030
+ "1009": "SCONJ|r-mark",
1031
+ "1010": "SCONJ|r-orphan",
1032
+ "1011": "SCONJ|root",
1033
+ "1012": "SYM",
1034
+ "1013": "SYM|l-dep",
1035
+ "1014": "SYM|r-clf",
1036
+ "1015": "SYM|r-nmod",
1037
+ "1016": "SYM|r-obj",
1038
+ "1017": "SYM|r-obl",
1039
+ "1018": "SYM|r-xcomp",
1040
+ "1019": "VERB",
1041
+ "1020": "VERB|Abbr=Yes",
1042
+ "1021": "VERB|Abbr=Yes|r-acl",
1043
+ "1022": "VERB|Foreign=Yes",
1044
+ "1023": "VERB|Foreign=Yes|l-nsubj",
1045
+ "1024": "VERB|Foreign=Yes|r-acl",
1046
+ "1025": "VERB|Foreign=Yes|r-advcl",
1047
+ "1026": "VERB|Foreign=Yes|r-ccomp",
1048
+ "1027": "VERB|Foreign=Yes|r-compound",
1049
+ "1028": "VERB|Foreign=Yes|r-conj",
1050
+ "1029": "VERB|Foreign=Yes|r-flat",
1051
+ "1030": "VERB|Foreign=Yes|r-nmod",
1052
+ "1031": "VERB|Foreign=Yes|r-xcomp",
1053
+ "1032": "VERB|Foreign=Yes|root",
1054
+ "1033": "VERB|Mood=Imp",
1055
+ "1034": "VERB|Mood=Imp|r-xcomp",
1056
+ "1035": "VERB|NounType=Class",
1057
+ "1036": "VERB|NounType=Class|r-acl",
1058
+ "1037": "VERB|NounType=Class|r-compound",
1059
+ "1038": "VERB|PartType=Adj",
1060
+ "1039": "VERB|PartType=Adj|r-acl",
1061
+ "1040": "VERB|Prefix=Yes",
1062
+ "1041": "VERB|Prefix=Yes|l-acl",
1063
+ "1042": "VERB|Prefix=Yes|l-nsubj",
1064
+ "1043": "VERB|Prefix=Yes|r-acl",
1065
+ "1044": "VERB|Prefix=Yes|r-advcl",
1066
+ "1045": "VERB|Prefix=Yes|r-ccomp",
1067
+ "1046": "VERB|Prefix=Yes|r-compound",
1068
+ "1047": "VERB|Prefix=Yes|r-conj",
1069
+ "1048": "VERB|Prefix=Yes|r-parataxis",
1070
+ "1049": "VERB|Prefix=Yes|root",
1071
+ "1050": "VERB|VerbType=Cop",
1072
+ "1051": "VERB|VerbType=Cop|l-advmod",
1073
+ "1052": "VERB|VerbType=Cop|l-cop",
1074
+ "1053": "VERB|VerbType=Cop|r-acl",
1075
+ "1054": "VERB|VerbType=Cop|r-advcl",
1076
+ "1055": "VERB|VerbType=Cop|r-ccomp",
1077
+ "1056": "VERB|VerbType=Cop|r-compound",
1078
+ "1057": "VERB|VerbType=Cop|r-parataxis",
1079
+ "1058": "VERB|VerbType=Cop|root",
1080
+ "1059": "VERB|Voice=Pass",
1081
+ "1060": "VERB|Voice=Pass|l-aux:pass",
1082
+ "1061": "VERB|l-acl",
1083
+ "1062": "VERB|l-advcl",
1084
+ "1063": "VERB|l-advmod",
1085
+ "1064": "VERB|l-aux",
1086
+ "1065": "VERB|l-case",
1087
+ "1066": "VERB|l-cc",
1088
+ "1067": "VERB|l-ccomp",
1089
+ "1068": "VERB|l-compound",
1090
+ "1069": "VERB|l-conj",
1091
+ "1070": "VERB|l-cop",
1092
+ "1071": "VERB|l-csubj",
1093
+ "1072": "VERB|l-discourse",
1094
+ "1073": "VERB|l-dislocated",
1095
+ "1074": "VERB|l-mark",
1096
+ "1075": "VERB|l-nsubj",
1097
+ "1076": "VERB|l-obl",
1098
+ "1077": "VERB|l-orphan",
1099
+ "1078": "VERB|l-xcomp",
1100
+ "1079": "VERB|r-acl",
1101
+ "1080": "VERB|r-acl:relcl",
1102
+ "1081": "VERB|r-advcl",
1103
+ "1082": "VERB|r-advmod",
1104
+ "1083": "VERB|r-appos",
1105
+ "1084": "VERB|r-aux",
1106
+ "1085": "VERB|r-case",
1107
+ "1086": "VERB|r-cc",
1108
+ "1087": "VERB|r-ccomp",
1109
+ "1088": "VERB|r-clf",
1110
+ "1089": "VERB|r-compound",
1111
+ "1090": "VERB|r-conj",
1112
+ "1091": "VERB|r-dep",
1113
+ "1092": "VERB|r-det",
1114
+ "1093": "VERB|r-discourse",
1115
+ "1094": "VERB|r-fixed",
1116
+ "1095": "VERB|r-flat",
1117
+ "1096": "VERB|r-list",
1118
+ "1097": "VERB|r-mark",
1119
+ "1098": "VERB|r-nmod",
1120
+ "1099": "VERB|r-nsubj",
1121
+ "1100": "VERB|r-obj",
1122
+ "1101": "VERB|r-obl",
1123
+ "1102": "VERB|r-orphan",
1124
+ "1103": "VERB|r-parataxis",
1125
+ "1104": "VERB|r-punct",
1126
+ "1105": "VERB|r-xcomp",
1127
+ "1106": "VERB|root",
1128
+ "1107": "X"
1129
+ },
1130
+ "initializer_range": 0.02,
1131
+ "label2id": {
1132
+ "ADP": 0,
1133
+ "ADP|Foreign=Yes": 1,
1134
+ "ADP|Foreign=Yes|l-case": 2,
1135
+ "ADP|NounType=Class": 3,
1136
+ "ADP|NounType=Class|l-case": 4,
1137
+ "ADP|Prefix=Yes": 5,
1138
+ "ADP|Prefix=Yes|l-case": 6,
1139
+ "ADP|Prefix=Yes|l-mark": 7,
1140
+ "ADP|l-acl": 8,
1141
+ "ADP|l-advcl": 9,
1142
+ "ADP|l-advmod": 10,
1143
+ "ADP|l-case": 11,
1144
+ "ADP|l-cc": 12,
1145
+ "ADP|l-dep": 13,
1146
+ "ADP|l-fixed": 14,
1147
+ "ADP|l-flat": 15,
1148
+ "ADP|l-mark": 16,
1149
+ "ADP|l-nmod": 17,
1150
+ "ADP|l-nsubj": 18,
1151
+ "ADP|l-obl": 19,
1152
+ "ADP|l-orphan": 20,
1153
+ "ADP|r-acl": 21,
1154
+ "ADP|r-advmod": 22,
1155
+ "ADP|r-case": 23,
1156
+ "ADP|r-compound": 24,
1157
+ "ADP|r-conj": 25,
1158
+ "ADP|r-fixed": 26,
1159
+ "ADP|r-flat": 27,
1160
+ "ADP|r-mark": 28,
1161
+ "ADP|r-obl": 29,
1162
+ "ADP|r-orphan": 30,
1163
+ "ADP|root": 31,
1164
+ "ADV": 32,
1165
+ "ADV|Foreign=Yes": 33,
1166
+ "ADV|Foreign=Yes|l-advmod": 34,
1167
+ "ADV|Foreign=Yes|r-advmod": 35,
1168
+ "ADV|NumType=Mult": 36,
1169
+ "ADV|NumType=Mult|r-advmod": 37,
1170
+ "ADV|PartType=Adv": 38,
1171
+ "ADV|PartType=Adv|l-advmod": 39,
1172
+ "ADV|PartType=Adv|l-mark": 40,
1173
+ "ADV|PartType=Adv|r-advmod": 41,
1174
+ "ADV|PartType=Enp": 42,
1175
+ "ADV|PartType=Enp|l-advmod": 43,
1176
+ "ADV|PartType=Enp|r-advmod": 44,
1177
+ "ADV|PartType=Int": 45,
1178
+ "ADV|PartType=Int|r-advmod": 46,
1179
+ "ADV|PartType=Int|r-fixed": 47,
1180
+ "ADV|Prefix=Yes": 48,
1181
+ "ADV|Prefix=Yes|l-advmod": 49,
1182
+ "ADV|Prefix=Yes|l-mark": 50,
1183
+ "ADV|Prefix=Yes|r-advmod": 51,
1184
+ "ADV|l-acl": 52,
1185
+ "ADV|l-advcl": 53,
1186
+ "ADV|l-advmod": 54,
1187
+ "ADV|l-aux": 55,
1188
+ "ADV|l-case": 56,
1189
+ "ADV|l-compound": 57,
1190
+ "ADV|l-dep": 58,
1191
+ "ADV|l-det": 59,
1192
+ "ADV|l-discourse": 60,
1193
+ "ADV|l-fixed": 61,
1194
+ "ADV|l-mark": 62,
1195
+ "ADV|l-orphan": 63,
1196
+ "ADV|r-acl": 64,
1197
+ "ADV|r-advcl": 65,
1198
+ "ADV|r-advmod": 66,
1199
+ "ADV|r-aux": 67,
1200
+ "ADV|r-ccomp": 68,
1201
+ "ADV|r-compound": 69,
1202
+ "ADV|r-conj": 70,
1203
+ "ADV|r-det": 71,
1204
+ "ADV|r-fixed": 72,
1205
+ "ADV|r-flat": 73,
1206
+ "ADV|r-mark": 74,
1207
+ "ADV|r-nmod": 75,
1208
+ "ADV|r-obj": 76,
1209
+ "ADV|r-orphan": 77,
1210
+ "ADV|r-xcomp": 78,
1211
+ "ADV|root": 79,
1212
+ "AUX": 80,
1213
+ "AUX|Foreign=Yes": 81,
1214
+ "AUX|Foreign=Yes|l-aux": 82,
1215
+ "AUX|NounType=Class": 83,
1216
+ "AUX|NounType=Class|r-appos": 84,
1217
+ "AUX|Prefix=Yes": 85,
1218
+ "AUX|Prefix=Yes|l-aux": 86,
1219
+ "AUX|Prefix=Yes|r-aux": 87,
1220
+ "AUX|VerbType=Cop": 88,
1221
+ "AUX|VerbType=Cop|l-acl": 89,
1222
+ "AUX|VerbType=Cop|l-advcl": 90,
1223
+ "AUX|VerbType=Cop|l-aux": 91,
1224
+ "AUX|VerbType=Cop|l-cop": 92,
1225
+ "AUX|VerbType=Cop|r-acl": 93,
1226
+ "AUX|VerbType=Cop|r-advcl": 94,
1227
+ "AUX|VerbType=Cop|r-aux": 95,
1228
+ "AUX|VerbType=Cop|r-conj": 96,
1229
+ "AUX|VerbType=Cop|r-mark": 97,
1230
+ "AUX|VerbType=Cop|root": 98,
1231
+ "AUX|l-advmod": 99,
1232
+ "AUX|l-aux": 100,
1233
+ "AUX|l-aux:pass": 101,
1234
+ "AUX|l-cop": 102,
1235
+ "AUX|l-mark": 103,
1236
+ "AUX|r-acl": 104,
1237
+ "AUX|r-advmod": 105,
1238
+ "AUX|r-aux": 106,
1239
+ "AUX|r-ccomp": 107,
1240
+ "AUX|r-clf": 108,
1241
+ "AUX|r-compound": 109,
1242
+ "AUX|r-conj": 110,
1243
+ "AUX|r-fixed": 111,
1244
+ "AUX|r-mark": 112,
1245
+ "AUX|root": 113,
1246
+ "B-ADP": 114,
1247
+ "B-ADP|Foreign=Yes": 115,
1248
+ "B-ADP|NounType=Class": 116,
1249
+ "B-ADP|Prefix=Yes": 117,
1250
+ "B-ADV": 118,
1251
+ "B-ADV|Foreign=Yes": 119,
1252
+ "B-ADV|NumType=Mult": 120,
1253
+ "B-ADV|PartType=Adv": 121,
1254
+ "B-ADV|PartType=Enp": 122,
1255
+ "B-ADV|PartType=Int": 123,
1256
+ "B-ADV|Prefix=Yes": 124,
1257
+ "B-AUX": 125,
1258
+ "B-AUX|Foreign=Yes": 126,
1259
+ "B-AUX|NounType=Class": 127,
1260
+ "B-AUX|Prefix=Yes": 128,
1261
+ "B-AUX|VerbType=Cop": 129,
1262
+ "B-CCONJ": 130,
1263
+ "B-CCONJ|Foreign=Yes": 131,
1264
+ "B-CCONJ|PronType=Prs": 132,
1265
+ "B-DET": 133,
1266
+ "B-DET|NumType=Mult": 134,
1267
+ "B-DET|PartType=Emp": 135,
1268
+ "B-DET|PartType=Int": 136,
1269
+ "B-DET|PronType=Int": 137,
1270
+ "B-INTJ": 138,
1271
+ "B-NOUN": 139,
1272
+ "B-NOUN|Abbr=Yes": 140,
1273
+ "B-NOUN|Abbr=Yes|Foreign=Yes": 141,
1274
+ "B-NOUN|Abbr=Yes|Prefix=Yes": 142,
1275
+ "B-NOUN|Foreign=Yes": 143,
1276
+ "B-NOUN|Foreign=Yes|NounType=Class": 144,
1277
+ "B-NOUN|Foreign=Yes|Prefix=Yes": 145,
1278
+ "B-NOUN|NameType=Com": 146,
1279
+ "B-NOUN|NameType=Geo": 147,
1280
+ "B-NOUN|NameType=Nat": 148,
1281
+ "B-NOUN|NameType=Oth": 149,
1282
+ "B-NOUN|NameType=Pro": 150,
1283
+ "B-NOUN|NameType=Prs": 151,
1284
+ "B-NOUN|NounType=Class": 152,
1285
+ "B-NOUN|NounType=Class|Prefix=Yes": 153,
1286
+ "B-NOUN|NumType=Mult": 154,
1287
+ "B-NOUN|PartType=Enp": 155,
1288
+ "B-NOUN|PartType=Int": 156,
1289
+ "B-NOUN|PartType=Res": 157,
1290
+ "B-NOUN|Prefix=Yes": 158,
1291
+ "B-NUM": 159,
1292
+ "B-NUM|Abbr=Yes": 160,
1293
+ "B-NUM|Foreign=Yes": 161,
1294
+ "B-NUM|NumType=Mult": 162,
1295
+ "B-NUM|Prefix=Yes": 163,
1296
+ "B-PART": 164,
1297
+ "B-PART|Aspect=Perf": 165,
1298
+ "B-PART|Aspect=Prog": 166,
1299
+ "B-PART|NameType=Oth": 167,
1300
+ "B-PART|NounType=Class|PartType=Emp": 168,
1301
+ "B-PART|NounType=Class|PartType=Emp|Prefix=Yes": 169,
1302
+ "B-PART|NounType=Class|Prefix=Yes": 170,
1303
+ "B-PART|NumType=Mult|PartType=Emp": 171,
1304
+ "B-PART|PartType=Adj": 172,
1305
+ "B-PART|PartType=Adv": 173,
1306
+ "B-PART|PartType=Emp": 174,
1307
+ "B-PART|PartType=Emp|Prefix=Yes": 175,
1308
+ "B-PART|PartType=Enp": 176,
1309
+ "B-PART|PartType=Int": 177,
1310
+ "B-PART|PartType=Neg": 178,
1311
+ "B-PART|PartType=Res": 179,
1312
+ "B-PART|Polarity=Neg": 180,
1313
+ "B-PART|Prefix=Yes": 181,
1314
+ "B-PART|PronType=Int": 182,
1315
+ "B-PRON": 183,
1316
+ "B-PRON|NounType=Class": 184,
1317
+ "B-PRON|Person=1": 185,
1318
+ "B-PRON|Person=2": 186,
1319
+ "B-PRON|Person=3": 187,
1320
+ "B-PRON|PronType=Int": 188,
1321
+ "B-PRON|PronType=Prs": 189,
1322
+ "B-PRON|PronType=Rcp": 190,
1323
+ "B-PROPN": 191,
1324
+ "B-PROPN|Abbr=Yes": 192,
1325
+ "B-PROPN|Abbr=Yes|Foreign=Yes|NameType=Oth": 193,
1326
+ "B-PROPN|Abbr=Yes|NameType=Com": 194,
1327
+ "B-PROPN|Foreign=Yes": 195,
1328
+ "B-PROPN|Foreign=Yes|NameType=Com": 196,
1329
+ "B-PROPN|Foreign=Yes|NameType=Geo": 197,
1330
+ "B-PROPN|Foreign=Yes|NameType=Giv": 198,
1331
+ "B-PROPN|Foreign=Yes|NameType=Oth": 199,
1332
+ "B-PROPN|Foreign=Yes|NameType=Prs": 200,
1333
+ "B-PROPN|Foreign=Yes|NameType=Sur": 201,
1334
+ "B-PROPN|NameType=Com": 202,
1335
+ "B-PROPN|NameType=Geo": 203,
1336
+ "B-PROPN|NameType=Giv": 204,
1337
+ "B-PROPN|NameType=Nat": 205,
1338
+ "B-PROPN|NameType=Oth": 206,
1339
+ "B-PROPN|NameType=Pro": 207,
1340
+ "B-PROPN|NameType=Prs": 208,
1341
+ "B-PROPN|NameType=Sur": 209,
1342
+ "B-PROPN|NounType=Class": 210,
1343
+ "B-PROPN|Prefix=Yes": 211,
1344
+ "B-PUNCT": 212,
1345
+ "B-PUNCT|NounType=Class": 213,
1346
+ "B-SCONJ": 214,
1347
+ "B-SCONJ|NumType=Mult": 215,
1348
+ "B-SCONJ|Prefix=Yes": 216,
1349
+ "B-SCONJ|VerbType=Cop": 217,
1350
+ "B-SYM": 218,
1351
+ "B-VERB": 219,
1352
+ "B-VERB|Abbr=Yes": 220,
1353
+ "B-VERB|Foreign=Yes": 221,
1354
+ "B-VERB|Mood=Imp": 222,
1355
+ "B-VERB|NounType=Class": 223,
1356
+ "B-VERB|PartType=Adj": 224,
1357
+ "B-VERB|Prefix=Yes": 225,
1358
+ "B-VERB|VerbType=Cop": 226,
1359
+ "B-VERB|Voice=Pass": 227,
1360
+ "B-X": 228,
1361
+ "CCONJ": 229,
1362
+ "CCONJ|Foreign=Yes": 230,
1363
+ "CCONJ|Foreign=Yes|l-cc": 231,
1364
+ "CCONJ|PronType=Prs": 232,
1365
+ "CCONJ|PronType=Prs|l-cc": 233,
1366
+ "CCONJ|l-advmod": 234,
1367
+ "CCONJ|l-case": 235,
1368
+ "CCONJ|l-cc": 236,
1369
+ "CCONJ|l-conj": 237,
1370
+ "CCONJ|l-discourse": 238,
1371
+ "CCONJ|l-fixed": 239,
1372
+ "CCONJ|l-flat": 240,
1373
+ "CCONJ|l-mark": 241,
1374
+ "CCONJ|l-nsubj": 242,
1375
+ "CCONJ|l-obj": 243,
1376
+ "CCONJ|l-orphan": 244,
1377
+ "CCONJ|r-cc": 245,
1378
+ "CCONJ|r-compound": 246,
1379
+ "CCONJ|r-fixed": 247,
1380
+ "CCONJ|r-mark": 248,
1381
+ "CCONJ|r-obl": 249,
1382
+ "CCONJ|root": 250,
1383
+ "DET": 251,
1384
+ "DET|NumType=Mult": 252,
1385
+ "DET|NumType=Mult|l-det": 253,
1386
+ "DET|PartType=Emp": 254,
1387
+ "DET|PartType=Emp|r-det": 255,
1388
+ "DET|PartType=Int": 256,
1389
+ "DET|PartType=Int|r-det": 257,
1390
+ "DET|PronType=Int": 258,
1391
+ "DET|l-advmod": 259,
1392
+ "DET|l-case": 260,
1393
+ "DET|l-compound": 261,
1394
+ "DET|l-det": 262,
1395
+ "DET|l-discourse": 263,
1396
+ "DET|l-nsubj": 264,
1397
+ "DET|l-obl": 265,
1398
+ "DET|l-orphan": 266,
1399
+ "DET|r-advmod": 267,
1400
+ "DET|r-compound": 268,
1401
+ "DET|r-conj": 269,
1402
+ "DET|r-dep": 270,
1403
+ "DET|r-det": 271,
1404
+ "DET|r-fixed": 272,
1405
+ "DET|r-flat": 273,
1406
+ "DET|r-list": 274,
1407
+ "DET|r-nmod": 275,
1408
+ "DET|r-nummod": 276,
1409
+ "DET|r-obj": 277,
1410
+ "DET|r-obl": 278,
1411
+ "DET|r-orphan": 279,
1412
+ "DET|root": 280,
1413
+ "I-ADP": 281,
1414
+ "I-ADP|Foreign=Yes": 282,
1415
+ "I-ADP|NounType=Class": 283,
1416
+ "I-ADP|Prefix=Yes": 284,
1417
+ "I-ADV": 285,
1418
+ "I-ADV|Foreign=Yes": 286,
1419
+ "I-ADV|NumType=Mult": 287,
1420
+ "I-ADV|PartType=Adv": 288,
1421
+ "I-ADV|PartType=Enp": 289,
1422
+ "I-ADV|PartType=Int": 290,
1423
+ "I-ADV|Prefix=Yes": 291,
1424
+ "I-AUX": 292,
1425
+ "I-AUX|Foreign=Yes": 293,
1426
+ "I-AUX|NounType=Class": 294,
1427
+ "I-AUX|Prefix=Yes": 295,
1428
+ "I-AUX|VerbType=Cop": 296,
1429
+ "I-CCONJ": 297,
1430
+ "I-CCONJ|Foreign=Yes": 298,
1431
+ "I-CCONJ|PronType=Prs": 299,
1432
+ "I-DET": 300,
1433
+ "I-DET|NumType=Mult": 301,
1434
+ "I-DET|PartType=Emp": 302,
1435
+ "I-DET|PartType=Int": 303,
1436
+ "I-DET|PronType=Int": 304,
1437
+ "I-INTJ": 305,
1438
+ "I-NOUN": 306,
1439
+ "I-NOUN|Abbr=Yes": 307,
1440
+ "I-NOUN|Abbr=Yes|Foreign=Yes": 308,
1441
+ "I-NOUN|Abbr=Yes|Prefix=Yes": 309,
1442
+ "I-NOUN|Foreign=Yes": 310,
1443
+ "I-NOUN|Foreign=Yes|NounType=Class": 311,
1444
+ "I-NOUN|Foreign=Yes|Prefix=Yes": 312,
1445
+ "I-NOUN|NameType=Com": 313,
1446
+ "I-NOUN|NameType=Geo": 314,
1447
+ "I-NOUN|NameType=Nat": 315,
1448
+ "I-NOUN|NameType=Oth": 316,
1449
+ "I-NOUN|NameType=Pro": 317,
1450
+ "I-NOUN|NameType=Prs": 318,
1451
+ "I-NOUN|NounType=Class": 319,
1452
+ "I-NOUN|NounType=Class|Prefix=Yes": 320,
1453
+ "I-NOUN|NumType=Mult": 321,
1454
+ "I-NOUN|PartType=Enp": 322,
1455
+ "I-NOUN|PartType=Int": 323,
1456
+ "I-NOUN|PartType=Res": 324,
1457
+ "I-NOUN|Prefix=Yes": 325,
1458
+ "I-NUM": 326,
1459
+ "I-NUM|Abbr=Yes": 327,
1460
+ "I-NUM|Foreign=Yes": 328,
1461
+ "I-NUM|NumType=Mult": 329,
1462
+ "I-NUM|Prefix=Yes": 330,
1463
+ "I-PART": 331,
1464
+ "I-PART|Aspect=Perf": 332,
1465
+ "I-PART|Aspect=Prog": 333,
1466
+ "I-PART|NameType=Oth": 334,
1467
+ "I-PART|NounType=Class|PartType=Emp": 335,
1468
+ "I-PART|NounType=Class|PartType=Emp|Prefix=Yes": 336,
1469
+ "I-PART|NounType=Class|Prefix=Yes": 337,
1470
+ "I-PART|NumType=Mult|PartType=Emp": 338,
1471
+ "I-PART|PartType=Adj": 339,
1472
+ "I-PART|PartType=Adv": 340,
1473
+ "I-PART|PartType=Emp": 341,
1474
+ "I-PART|PartType=Emp|Prefix=Yes": 342,
1475
+ "I-PART|PartType=Enp": 343,
1476
+ "I-PART|PartType=Int": 344,
1477
+ "I-PART|PartType=Neg": 345,
1478
+ "I-PART|PartType=Res": 346,
1479
+ "I-PART|Polarity=Neg": 347,
1480
+ "I-PART|Prefix=Yes": 348,
1481
+ "I-PART|PronType=Int": 349,
1482
+ "I-PRON": 350,
1483
+ "I-PRON|NounType=Class": 351,
1484
+ "I-PRON|Person=1": 352,
1485
+ "I-PRON|Person=2": 353,
1486
+ "I-PRON|Person=3": 354,
1487
+ "I-PRON|PronType=Int": 355,
1488
+ "I-PRON|PronType=Prs": 356,
1489
+ "I-PRON|PronType=Rcp": 357,
1490
+ "I-PROPN": 358,
1491
+ "I-PROPN|Abbr=Yes": 359,
1492
+ "I-PROPN|Abbr=Yes|Foreign=Yes|NameType=Oth": 360,
1493
+ "I-PROPN|Abbr=Yes|NameType=Com": 361,
1494
+ "I-PROPN|Foreign=Yes": 362,
1495
+ "I-PROPN|Foreign=Yes|NameType=Com": 363,
1496
+ "I-PROPN|Foreign=Yes|NameType=Geo": 364,
1497
+ "I-PROPN|Foreign=Yes|NameType=Giv": 365,
1498
+ "I-PROPN|Foreign=Yes|NameType=Oth": 366,
1499
+ "I-PROPN|Foreign=Yes|NameType=Prs": 367,
1500
+ "I-PROPN|Foreign=Yes|NameType=Sur": 368,
1501
+ "I-PROPN|NameType=Com": 369,
1502
+ "I-PROPN|NameType=Geo": 370,
1503
+ "I-PROPN|NameType=Giv": 371,
1504
+ "I-PROPN|NameType=Nat": 372,
1505
+ "I-PROPN|NameType=Oth": 373,
1506
+ "I-PROPN|NameType=Pro": 374,
1507
+ "I-PROPN|NameType=Prs": 375,
1508
+ "I-PROPN|NameType=Sur": 376,
1509
+ "I-PROPN|NounType=Class": 377,
1510
+ "I-PROPN|Prefix=Yes": 378,
1511
+ "I-PUNCT": 379,
1512
+ "I-PUNCT|NounType=Class": 380,
1513
+ "I-SCONJ": 381,
1514
+ "I-SCONJ|NumType=Mult": 382,
1515
+ "I-SCONJ|Prefix=Yes": 383,
1516
+ "I-SCONJ|VerbType=Cop": 384,
1517
+ "I-SYM": 385,
1518
+ "I-VERB": 386,
1519
+ "I-VERB|Abbr=Yes": 387,
1520
+ "I-VERB|Foreign=Yes": 388,
1521
+ "I-VERB|Mood=Imp": 389,
1522
+ "I-VERB|NounType=Class": 390,
1523
+ "I-VERB|PartType=Adj": 391,
1524
+ "I-VERB|Prefix=Yes": 392,
1525
+ "I-VERB|VerbType=Cop": 393,
1526
+ "I-VERB|Voice=Pass": 394,
1527
+ "I-X": 395,
1528
+ "INTJ": 396,
1529
+ "INTJ|l-nsubj": 397,
1530
+ "INTJ|r-acl": 398,
1531
+ "INTJ|root": 399,
1532
+ "NOUN": 400,
1533
+ "NOUN|Abbr=Yes": 401,
1534
+ "NOUN|Abbr=Yes|Foreign=Yes": 402,
1535
+ "NOUN|Abbr=Yes|Foreign=Yes|r-nmod": 403,
1536
+ "NOUN|Abbr=Yes|Prefix=Yes": 404,
1537
+ "NOUN|Abbr=Yes|Prefix=Yes|l-flat": 405,
1538
+ "NOUN|Abbr=Yes|l-flat": 406,
1539
+ "NOUN|Abbr=Yes|l-nmod": 407,
1540
+ "NOUN|Abbr=Yes|l-nsubj": 408,
1541
+ "NOUN|Abbr=Yes|l-obl": 409,
1542
+ "NOUN|Abbr=Yes|r-acl": 410,
1543
+ "NOUN|Abbr=Yes|r-appos": 411,
1544
+ "NOUN|Abbr=Yes|r-clf": 412,
1545
+ "NOUN|Abbr=Yes|r-conj": 413,
1546
+ "NOUN|Abbr=Yes|r-fixed": 414,
1547
+ "NOUN|Abbr=Yes|r-flat": 415,
1548
+ "NOUN|Abbr=Yes|r-nmod": 416,
1549
+ "NOUN|Abbr=Yes|r-obj": 417,
1550
+ "NOUN|Abbr=Yes|r-obl": 418,
1551
+ "NOUN|Foreign=Yes": 419,
1552
+ "NOUN|Foreign=Yes|NounType=Class": 420,
1553
+ "NOUN|Foreign=Yes|NounType=Class|r-clf": 421,
1554
+ "NOUN|Foreign=Yes|NounType=Class|r-obj": 422,
1555
+ "NOUN|Foreign=Yes|Prefix=Yes": 423,
1556
+ "NOUN|Foreign=Yes|Prefix=Yes|l-flat": 424,
1557
+ "NOUN|Foreign=Yes|Prefix=Yes|r-appos": 425,
1558
+ "NOUN|Foreign=Yes|l-dislocated": 426,
1559
+ "NOUN|Foreign=Yes|l-flat": 427,
1560
+ "NOUN|Foreign=Yes|l-nmod": 428,
1561
+ "NOUN|Foreign=Yes|l-nsubj": 429,
1562
+ "NOUN|Foreign=Yes|l-obl": 430,
1563
+ "NOUN|Foreign=Yes|r-acl": 431,
1564
+ "NOUN|Foreign=Yes|r-advcl": 432,
1565
+ "NOUN|Foreign=Yes|r-advmod": 433,
1566
+ "NOUN|Foreign=Yes|r-appos": 434,
1567
+ "NOUN|Foreign=Yes|r-ccomp": 435,
1568
+ "NOUN|Foreign=Yes|r-clf": 436,
1569
+ "NOUN|Foreign=Yes|r-compound": 437,
1570
+ "NOUN|Foreign=Yes|r-conj": 438,
1571
+ "NOUN|Foreign=Yes|r-flat": 439,
1572
+ "NOUN|Foreign=Yes|r-iobj": 440,
1573
+ "NOUN|Foreign=Yes|r-list": 441,
1574
+ "NOUN|Foreign=Yes|r-nmod": 442,
1575
+ "NOUN|Foreign=Yes|r-obj": 443,
1576
+ "NOUN|Foreign=Yes|r-obl": 444,
1577
+ "NOUN|Foreign=Yes|r-xcomp": 445,
1578
+ "NOUN|Foreign=Yes|root": 446,
1579
+ "NOUN|NameType=Com": 447,
1580
+ "NOUN|NameType=Com|r-nmod": 448,
1581
+ "NOUN|NameType=Geo": 449,
1582
+ "NOUN|NameType=Geo|l-nsubj": 450,
1583
+ "NOUN|NameType=Geo|r-nmod": 451,
1584
+ "NOUN|NameType=Geo|r-obj": 452,
1585
+ "NOUN|NameType=Nat": 453,
1586
+ "NOUN|NameType=Nat|r-nmod": 454,
1587
+ "NOUN|NameType=Oth": 455,
1588
+ "NOUN|NameType=Oth|l-nsubj": 456,
1589
+ "NOUN|NameType=Oth|r-conj": 457,
1590
+ "NOUN|NameType=Oth|r-flat": 458,
1591
+ "NOUN|NameType=Oth|r-nmod": 459,
1592
+ "NOUN|NameType=Pro": 460,
1593
+ "NOUN|NameType=Pro|r-nmod": 461,
1594
+ "NOUN|NameType=Prs": 462,
1595
+ "NOUN|NameType=Prs|l-nsubj": 463,
1596
+ "NOUN|NameType=Prs|r-nmod": 464,
1597
+ "NOUN|NounType=Class": 465,
1598
+ "NOUN|NounType=Class|Prefix=Yes": 466,
1599
+ "NOUN|NounType=Class|Prefix=Yes|l-advcl": 467,
1600
+ "NOUN|NounType=Class|Prefix=Yes|l-advmod": 468,
1601
+ "NOUN|NounType=Class|Prefix=Yes|l-mark": 469,
1602
+ "NOUN|NounType=Class|Prefix=Yes|l-nmod": 470,
1603
+ "NOUN|NounType=Class|Prefix=Yes|l-nsubj": 471,
1604
+ "NOUN|NounType=Class|Prefix=Yes|r-advcl": 472,
1605
+ "NOUN|NounType=Class|Prefix=Yes|r-clf": 473,
1606
+ "NOUN|NounType=Class|Prefix=Yes|r-nmod": 474,
1607
+ "NOUN|NounType=Class|Prefix=Yes|r-obj": 475,
1608
+ "NOUN|NounType=Class|l-advcl": 476,
1609
+ "NOUN|NounType=Class|l-advmod": 477,
1610
+ "NOUN|NounType=Class|l-clf": 478,
1611
+ "NOUN|NounType=Class|l-dislocated": 479,
1612
+ "NOUN|NounType=Class|l-nmod": 480,
1613
+ "NOUN|NounType=Class|l-nsubj": 481,
1614
+ "NOUN|NounType=Class|l-obj": 482,
1615
+ "NOUN|NounType=Class|l-obl": 483,
1616
+ "NOUN|NounType=Class|r-acl": 484,
1617
+ "NOUN|NounType=Class|r-advcl": 485,
1618
+ "NOUN|NounType=Class|r-advmod": 486,
1619
+ "NOUN|NounType=Class|r-appos": 487,
1620
+ "NOUN|NounType=Class|r-cc": 488,
1621
+ "NOUN|NounType=Class|r-ccomp": 489,
1622
+ "NOUN|NounType=Class|r-clf": 490,
1623
+ "NOUN|NounType=Class|r-compound": 491,
1624
+ "NOUN|NounType=Class|r-conj": 492,
1625
+ "NOUN|NounType=Class|r-dislocated": 493,
1626
+ "NOUN|NounType=Class|r-fixed": 494,
1627
+ "NOUN|NounType=Class|r-flat": 495,
1628
+ "NOUN|NounType=Class|r-iobj": 496,
1629
+ "NOUN|NounType=Class|r-list": 497,
1630
+ "NOUN|NounType=Class|r-nmod": 498,
1631
+ "NOUN|NounType=Class|r-nummod": 499,
1632
+ "NOUN|NounType=Class|r-obj": 500,
1633
+ "NOUN|NounType=Class|r-obl": 501,
1634
+ "NOUN|NounType=Class|r-orphan": 502,
1635
+ "NOUN|NounType=Class|r-xcomp": 503,
1636
+ "NOUN|NounType=Class|root": 504,
1637
+ "NOUN|NumType=Mult": 505,
1638
+ "NOUN|NumType=Mult|r-advcl": 506,
1639
+ "NOUN|NumType=Mult|r-nmod": 507,
1640
+ "NOUN|NumType=Mult|r-obj": 508,
1641
+ "NOUN|PartType=Enp": 509,
1642
+ "NOUN|PartType=Enp|r-obj": 510,
1643
+ "NOUN|PartType=Enp|r-obl": 511,
1644
+ "NOUN|PartType=Int": 512,
1645
+ "NOUN|PartType=Int|r-obj": 513,
1646
+ "NOUN|PartType=Res": 514,
1647
+ "NOUN|PartType=Res|r-nmod": 515,
1648
+ "NOUN|PartType=Res|r-obj": 516,
1649
+ "NOUN|Prefix=Yes": 517,
1650
+ "NOUN|Prefix=Yes|l-acl": 518,
1651
+ "NOUN|Prefix=Yes|l-advcl": 519,
1652
+ "NOUN|Prefix=Yes|l-clf": 520,
1653
+ "NOUN|Prefix=Yes|l-csubj": 521,
1654
+ "NOUN|Prefix=Yes|l-dislocated": 522,
1655
+ "NOUN|Prefix=Yes|l-flat": 523,
1656
+ "NOUN|Prefix=Yes|l-nmod": 524,
1657
+ "NOUN|Prefix=Yes|l-nsubj": 525,
1658
+ "NOUN|Prefix=Yes|l-obj": 526,
1659
+ "NOUN|Prefix=Yes|l-obl": 527,
1660
+ "NOUN|Prefix=Yes|r-acl": 528,
1661
+ "NOUN|Prefix=Yes|r-advcl": 529,
1662
+ "NOUN|Prefix=Yes|r-advmod": 530,
1663
+ "NOUN|Prefix=Yes|r-appos": 531,
1664
+ "NOUN|Prefix=Yes|r-case": 532,
1665
+ "NOUN|Prefix=Yes|r-cc": 533,
1666
+ "NOUN|Prefix=Yes|r-ccomp": 534,
1667
+ "NOUN|Prefix=Yes|r-clf": 535,
1668
+ "NOUN|Prefix=Yes|r-compound": 536,
1669
+ "NOUN|Prefix=Yes|r-conj": 537,
1670
+ "NOUN|Prefix=Yes|r-dislocated": 538,
1671
+ "NOUN|Prefix=Yes|r-fixed": 539,
1672
+ "NOUN|Prefix=Yes|r-flat": 540,
1673
+ "NOUN|Prefix=Yes|r-iobj": 541,
1674
+ "NOUN|Prefix=Yes|r-list": 542,
1675
+ "NOUN|Prefix=Yes|r-nmod": 543,
1676
+ "NOUN|Prefix=Yes|r-nummod": 544,
1677
+ "NOUN|Prefix=Yes|r-obj": 545,
1678
+ "NOUN|Prefix=Yes|r-obl": 546,
1679
+ "NOUN|Prefix=Yes|r-orphan": 547,
1680
+ "NOUN|Prefix=Yes|r-xcomp": 548,
1681
+ "NOUN|Prefix=Yes|root": 549,
1682
+ "NOUN|l-acl": 550,
1683
+ "NOUN|l-advcl": 551,
1684
+ "NOUN|l-advmod": 552,
1685
+ "NOUN|l-aux": 553,
1686
+ "NOUN|l-case": 554,
1687
+ "NOUN|l-ccomp": 555,
1688
+ "NOUN|l-compound": 556,
1689
+ "NOUN|l-csubj": 557,
1690
+ "NOUN|l-discourse": 558,
1691
+ "NOUN|l-dislocated": 559,
1692
+ "NOUN|l-expl": 560,
1693
+ "NOUN|l-flat": 561,
1694
+ "NOUN|l-iobj": 562,
1695
+ "NOUN|l-mark": 563,
1696
+ "NOUN|l-nmod": 564,
1697
+ "NOUN|l-nsubj": 565,
1698
+ "NOUN|l-nummod": 566,
1699
+ "NOUN|l-obj": 567,
1700
+ "NOUN|l-obl": 568,
1701
+ "NOUN|l-orphan": 569,
1702
+ "NOUN|l-vocative": 570,
1703
+ "NOUN|r-acl": 571,
1704
+ "NOUN|r-advcl": 572,
1705
+ "NOUN|r-advmod": 573,
1706
+ "NOUN|r-appos": 574,
1707
+ "NOUN|r-case": 575,
1708
+ "NOUN|r-cc": 576,
1709
+ "NOUN|r-ccomp": 577,
1710
+ "NOUN|r-clf": 578,
1711
+ "NOUN|r-compound": 579,
1712
+ "NOUN|r-conj": 580,
1713
+ "NOUN|r-cop": 581,
1714
+ "NOUN|r-discourse": 582,
1715
+ "NOUN|r-dislocated": 583,
1716
+ "NOUN|r-fixed": 584,
1717
+ "NOUN|r-flat": 585,
1718
+ "NOUN|r-flat:name": 586,
1719
+ "NOUN|r-iobj": 587,
1720
+ "NOUN|r-list": 588,
1721
+ "NOUN|r-mark": 589,
1722
+ "NOUN|r-nmod": 590,
1723
+ "NOUN|r-nmod:poss": 591,
1724
+ "NOUN|r-nsubj": 592,
1725
+ "NOUN|r-nummod": 593,
1726
+ "NOUN|r-obj": 594,
1727
+ "NOUN|r-obl": 595,
1728
+ "NOUN|r-obl:tmod": 596,
1729
+ "NOUN|r-orphan": 597,
1730
+ "NOUN|r-parataxis": 598,
1731
+ "NOUN|r-xcomp": 599,
1732
+ "NOUN|root": 600,
1733
+ "NUM": 601,
1734
+ "NUM|Abbr=Yes": 602,
1735
+ "NUM|Abbr=Yes|r-flat": 603,
1736
+ "NUM|Abbr=Yes|r-nummod": 604,
1737
+ "NUM|Abbr=Yes|r-obj": 605,
1738
+ "NUM|Foreign=Yes": 606,
1739
+ "NUM|Foreign=Yes|r-clf": 607,
1740
+ "NUM|NumType=Mult": 608,
1741
+ "NUM|NumType=Mult|l-advmod": 609,
1742
+ "NUM|NumType=Mult|l-nummod": 610,
1743
+ "NUM|NumType=Mult|r-advmod": 611,
1744
+ "NUM|Prefix=Yes": 612,
1745
+ "NUM|Prefix=Yes|l-nummod": 613,
1746
+ "NUM|l-advcl": 614,
1747
+ "NUM|l-advmod": 615,
1748
+ "NUM|l-case": 616,
1749
+ "NUM|l-clf": 617,
1750
+ "NUM|l-dep": 618,
1751
+ "NUM|l-flat": 619,
1752
+ "NUM|l-nmod": 620,
1753
+ "NUM|l-nsubj": 621,
1754
+ "NUM|l-nummod": 622,
1755
+ "NUM|l-obl": 623,
1756
+ "NUM|r-acl": 624,
1757
+ "NUM|r-advmod": 625,
1758
+ "NUM|r-appos": 626,
1759
+ "NUM|r-ccomp": 627,
1760
+ "NUM|r-clf": 628,
1761
+ "NUM|r-compound": 629,
1762
+ "NUM|r-conj": 630,
1763
+ "NUM|r-det": 631,
1764
+ "NUM|r-fixed": 632,
1765
+ "NUM|r-flat": 633,
1766
+ "NUM|r-flat:name": 634,
1767
+ "NUM|r-iobj": 635,
1768
+ "NUM|r-nmod": 636,
1769
+ "NUM|r-nummod": 637,
1770
+ "NUM|r-obj": 638,
1771
+ "NUM|r-obl": 639,
1772
+ "NUM|r-obl:tmod": 640,
1773
+ "NUM|root": 641,
1774
+ "PART": 642,
1775
+ "PART|Aspect=Perf": 643,
1776
+ "PART|Aspect=Prog": 644,
1777
+ "PART|NameType=Oth": 645,
1778
+ "PART|NameType=Oth|l-advmod": 646,
1779
+ "PART|NounType=Class|PartType=Emp": 647,
1780
+ "PART|NounType=Class|PartType=Emp|Prefix=Yes": 648,
1781
+ "PART|NounType=Class|PartType=Emp|Prefix=Yes|l-mark": 649,
1782
+ "PART|NounType=Class|PartType=Emp|l-mark": 650,
1783
+ "PART|NounType=Class|Prefix=Yes": 651,
1784
+ "PART|NounType=Class|Prefix=Yes|l-mark": 652,
1785
+ "PART|NumType=Mult|PartType=Emp": 653,
1786
+ "PART|NumType=Mult|PartType=Emp|l-mark": 654,
1787
+ "PART|PartType=Adj": 655,
1788
+ "PART|PartType=Adj|l-mark": 656,
1789
+ "PART|PartType=Adj|l-orphan": 657,
1790
+ "PART|PartType=Adj|r-acl": 658,
1791
+ "PART|PartType=Adj|r-compound": 659,
1792
+ "PART|PartType=Adj|r-nmod": 660,
1793
+ "PART|PartType=Adv": 661,
1794
+ "PART|PartType=Adv|l-advmod": 662,
1795
+ "PART|PartType=Adv|l-mark": 663,
1796
+ "PART|PartType=Adv|r-advmod": 664,
1797
+ "PART|PartType=Emp": 665,
1798
+ "PART|PartType=Emp|Prefix=Yes": 666,
1799
+ "PART|PartType=Emp|Prefix=Yes|l-advmod": 667,
1800
+ "PART|PartType=Emp|Prefix=Yes|l-aux": 668,
1801
+ "PART|PartType=Emp|Prefix=Yes|l-mark": 669,
1802
+ "PART|PartType=Emp|l-advmod": 670,
1803
+ "PART|PartType=Emp|l-case": 671,
1804
+ "PART|PartType=Emp|l-discourse": 672,
1805
+ "PART|PartType=Emp|l-mark": 673,
1806
+ "PART|PartType=Emp|r-acl": 674,
1807
+ "PART|PartType=Emp|r-advmod": 675,
1808
+ "PART|PartType=Emp|r-aux": 676,
1809
+ "PART|PartType=Emp|r-compound": 677,
1810
+ "PART|PartType=Emp|r-det": 678,
1811
+ "PART|PartType=Emp|r-fixed": 679,
1812
+ "PART|PartType=Emp|r-mark": 680,
1813
+ "PART|PartType=Emp|r-nmod": 681,
1814
+ "PART|PartType=Enp": 682,
1815
+ "PART|PartType=Enp|l-discourse": 683,
1816
+ "PART|PartType=Enp|r-acl": 684,
1817
+ "PART|PartType=Enp|r-advmod": 685,
1818
+ "PART|PartType=Enp|r-compound": 686,
1819
+ "PART|PartType=Enp|r-dep": 687,
1820
+ "PART|PartType=Enp|r-det": 688,
1821
+ "PART|PartType=Enp|r-discourse": 689,
1822
+ "PART|PartType=Enp|r-fixed": 690,
1823
+ "PART|PartType=Enp|r-obl": 691,
1824
+ "PART|PartType=Int": 692,
1825
+ "PART|PartType=Int|l-advmod": 693,
1826
+ "PART|PartType=Int|l-mark": 694,
1827
+ "PART|PartType=Int|r-acl": 695,
1828
+ "PART|PartType=Int|r-advmod": 696,
1829
+ "PART|PartType=Int|r-dep": 697,
1830
+ "PART|PartType=Int|r-discourse": 698,
1831
+ "PART|PartType=Int|r-nmod": 699,
1832
+ "PART|PartType=Int|r-obj": 700,
1833
+ "PART|PartType=Int|r-obl": 701,
1834
+ "PART|PartType=Neg": 702,
1835
+ "PART|PartType=Neg|l-advcl": 703,
1836
+ "PART|PartType=Neg|l-advmod": 704,
1837
+ "PART|PartType=Neg|l-aux": 705,
1838
+ "PART|PartType=Neg|l-mark": 706,
1839
+ "PART|PartType=Neg|r-acl": 707,
1840
+ "PART|PartType=Neg|r-advmod": 708,
1841
+ "PART|PartType=Neg|r-fixed": 709,
1842
+ "PART|PartType=Res": 710,
1843
+ "PART|PartType=Res|r-advmod": 711,
1844
+ "PART|PartType=Res|r-discourse": 712,
1845
+ "PART|PartType=Res|r-fixed": 713,
1846
+ "PART|Polarity=Neg": 714,
1847
+ "PART|Polarity=Neg|l-advmod": 715,
1848
+ "PART|Prefix=Yes": 716,
1849
+ "PART|Prefix=Yes|l-advmod": 717,
1850
+ "PART|Prefix=Yes|l-aux": 718,
1851
+ "PART|Prefix=Yes|l-mark": 719,
1852
+ "PART|Prefix=Yes|r-acl": 720,
1853
+ "PART|Prefix=Yes|r-nmod": 721,
1854
+ "PART|PronType=Int": 722,
1855
+ "PART|PronType=Int|r-acl": 723,
1856
+ "PART|PronType=Int|r-advmod": 724,
1857
+ "PART|PronType=Int|r-obj": 725,
1858
+ "PART|PronType=Int|root": 726,
1859
+ "PART|l-advmod": 727,
1860
+ "PART|l-cc": 728,
1861
+ "PART|l-discourse": 729,
1862
+ "PART|l-mark": 730,
1863
+ "PART|l-nsubj": 731,
1864
+ "PART|r-acl": 732,
1865
+ "PART|r-advmod": 733,
1866
+ "PART|r-ccomp": 734,
1867
+ "PART|r-clf": 735,
1868
+ "PART|r-compound": 736,
1869
+ "PART|r-conj": 737,
1870
+ "PART|r-discourse": 738,
1871
+ "PART|r-fixed": 739,
1872
+ "PART|r-mark": 740,
1873
+ "PART|r-nmod": 741,
1874
+ "PART|r-nmod:poss": 742,
1875
+ "PART|r-obj": 743,
1876
+ "PART|r-obl": 744,
1877
+ "PART|root": 745,
1878
+ "PRON": 746,
1879
+ "PRON|NounType=Class": 747,
1880
+ "PRON|NounType=Class|r-clf": 748,
1881
+ "PRON|Person=1": 749,
1882
+ "PRON|Person=1|l-nsubj": 750,
1883
+ "PRON|Person=1|r-compound": 751,
1884
+ "PRON|Person=1|r-nmod:poss": 752,
1885
+ "PRON|Person=2": 753,
1886
+ "PRON|Person=2|r-compound": 754,
1887
+ "PRON|Person=3": 755,
1888
+ "PRON|Person=3|l-nsubj": 756,
1889
+ "PRON|Person=3|r-appos": 757,
1890
+ "PRON|Person=3|r-compound": 758,
1891
+ "PRON|Person=3|r-nmod:poss": 759,
1892
+ "PRON|Person=3|r-obj": 760,
1893
+ "PRON|PronType=Int": 761,
1894
+ "PRON|PronType=Int|l-nsubj": 762,
1895
+ "PRON|PronType=Int|r-obj": 763,
1896
+ "PRON|PronType=Int|r-obl": 764,
1897
+ "PRON|PronType=Prs": 765,
1898
+ "PRON|PronType=Prs|l-advmod": 766,
1899
+ "PRON|PronType=Prs|l-expl": 767,
1900
+ "PRON|PronType=Prs|l-nsubj": 768,
1901
+ "PRON|PronType=Prs|l-obj": 769,
1902
+ "PRON|PronType=Prs|l-obl": 770,
1903
+ "PRON|PronType=Prs|r-advcl": 771,
1904
+ "PRON|PronType=Prs|r-advmod": 772,
1905
+ "PRON|PronType=Prs|r-ccomp": 773,
1906
+ "PRON|PronType=Prs|r-clf": 774,
1907
+ "PRON|PronType=Prs|r-conj": 775,
1908
+ "PRON|PronType=Prs|r-nmod": 776,
1909
+ "PRON|PronType=Prs|r-nsubj": 777,
1910
+ "PRON|PronType=Prs|r-obj": 778,
1911
+ "PRON|PronType=Prs|r-obl": 779,
1912
+ "PRON|PronType=Prs|root": 780,
1913
+ "PRON|PronType=Rcp": 781,
1914
+ "PRON|PronType=Rcp|r-advmod": 782,
1915
+ "PRON|PronType=Rcp|r-iobj": 783,
1916
+ "PRON|PronType=Rcp|r-nmod": 784,
1917
+ "PRON|PronType=Rcp|r-obj": 785,
1918
+ "PRON|PronType=Rcp|r-obl": 786,
1919
+ "PRON|l-advcl": 787,
1920
+ "PRON|l-advmod": 788,
1921
+ "PRON|l-compound": 789,
1922
+ "PRON|l-csubj": 790,
1923
+ "PRON|l-dislocated": 791,
1924
+ "PRON|l-expl": 792,
1925
+ "PRON|l-iobj": 793,
1926
+ "PRON|l-mark": 794,
1927
+ "PRON|l-nsubj": 795,
1928
+ "PRON|l-obj": 796,
1929
+ "PRON|l-obl": 797,
1930
+ "PRON|r-acl": 798,
1931
+ "PRON|r-advmod": 799,
1932
+ "PRON|r-appos": 800,
1933
+ "PRON|r-ccomp": 801,
1934
+ "PRON|r-compound": 802,
1935
+ "PRON|r-conj": 803,
1936
+ "PRON|r-det": 804,
1937
+ "PRON|r-discourse": 805,
1938
+ "PRON|r-fixed": 806,
1939
+ "PRON|r-flat": 807,
1940
+ "PRON|r-iobj": 808,
1941
+ "PRON|r-nmod": 809,
1942
+ "PRON|r-nsubj": 810,
1943
+ "PRON|r-obj": 811,
1944
+ "PRON|r-obl": 812,
1945
+ "PRON|r-xcomp": 813,
1946
+ "PROPN": 814,
1947
+ "PROPN|Abbr=Yes": 815,
1948
+ "PROPN|Abbr=Yes|Foreign=Yes|NameType=Oth": 816,
1949
+ "PROPN|Abbr=Yes|Foreign=Yes|NameType=Oth|r-obj": 817,
1950
+ "PROPN|Abbr=Yes|NameType=Com": 818,
1951
+ "PROPN|Abbr=Yes|NameType=Com|r-advmod": 819,
1952
+ "PROPN|Abbr=Yes|NameType=Com|r-nmod": 820,
1953
+ "PROPN|Abbr=Yes|l-nmod": 821,
1954
+ "PROPN|Abbr=Yes|l-nsubj": 822,
1955
+ "PROPN|Abbr=Yes|r-nmod": 823,
1956
+ "PROPN|Foreign=Yes": 824,
1957
+ "PROPN|Foreign=Yes|NameType=Com": 825,
1958
+ "PROPN|Foreign=Yes|NameType=Com|l-nsubj": 826,
1959
+ "PROPN|Foreign=Yes|NameType=Com|r-list": 827,
1960
+ "PROPN|Foreign=Yes|NameType=Com|r-nmod": 828,
1961
+ "PROPN|Foreign=Yes|NameType=Com|r-obl": 829,
1962
+ "PROPN|Foreign=Yes|NameType=Geo": 830,
1963
+ "PROPN|Foreign=Yes|NameType=Geo|r-obj": 831,
1964
+ "PROPN|Foreign=Yes|NameType=Geo|r-obl": 832,
1965
+ "PROPN|Foreign=Yes|NameType=Giv": 833,
1966
+ "PROPN|Foreign=Yes|NameType=Giv|l-nsubj": 834,
1967
+ "PROPN|Foreign=Yes|NameType=Oth": 835,
1968
+ "PROPN|Foreign=Yes|NameType=Oth|r-conj": 836,
1969
+ "PROPN|Foreign=Yes|NameType=Oth|r-flat": 837,
1970
+ "PROPN|Foreign=Yes|NameType=Oth|r-nmod": 838,
1971
+ "PROPN|Foreign=Yes|NameType=Prs": 839,
1972
+ "PROPN|Foreign=Yes|NameType=Prs|l-flat": 840,
1973
+ "PROPN|Foreign=Yes|NameType=Prs|l-nsubj": 841,
1974
+ "PROPN|Foreign=Yes|NameType=Prs|r-conj": 842,
1975
+ "PROPN|Foreign=Yes|NameType=Prs|r-flat": 843,
1976
+ "PROPN|Foreign=Yes|NameType=Prs|r-nmod": 844,
1977
+ "PROPN|Foreign=Yes|NameType=Prs|r-obj": 845,
1978
+ "PROPN|Foreign=Yes|NameType=Prs|r-obl": 846,
1979
+ "PROPN|Foreign=Yes|NameType=Sur": 847,
1980
+ "PROPN|Foreign=Yes|NameType=Sur|r-flat": 848,
1981
+ "PROPN|Foreign=Yes|l-flat": 849,
1982
+ "PROPN|Foreign=Yes|l-nmod": 850,
1983
+ "PROPN|Foreign=Yes|l-nsubj": 851,
1984
+ "PROPN|Foreign=Yes|l-obl": 852,
1985
+ "PROPN|Foreign=Yes|r-appos": 853,
1986
+ "PROPN|Foreign=Yes|r-ccomp": 854,
1987
+ "PROPN|Foreign=Yes|r-compound": 855,
1988
+ "PROPN|Foreign=Yes|r-conj": 856,
1989
+ "PROPN|Foreign=Yes|r-flat": 857,
1990
+ "PROPN|Foreign=Yes|r-iobj": 858,
1991
+ "PROPN|Foreign=Yes|r-list": 859,
1992
+ "PROPN|Foreign=Yes|r-nmod": 860,
1993
+ "PROPN|Foreign=Yes|r-nsubj": 861,
1994
+ "PROPN|Foreign=Yes|r-obj": 862,
1995
+ "PROPN|Foreign=Yes|r-obl": 863,
1996
+ "PROPN|Foreign=Yes|root": 864,
1997
+ "PROPN|NameType=Com": 865,
1998
+ "PROPN|NameType=Com|l-nsubj": 866,
1999
+ "PROPN|NameType=Com|l-obl": 867,
2000
+ "PROPN|NameType=Com|r-appos": 868,
2001
+ "PROPN|NameType=Com|r-conj": 869,
2002
+ "PROPN|NameType=Com|r-flat": 870,
2003
+ "PROPN|NameType=Com|r-list": 871,
2004
+ "PROPN|NameType=Com|r-nmod": 872,
2005
+ "PROPN|NameType=Com|r-nsubj": 873,
2006
+ "PROPN|NameType=Com|r-obj": 874,
2007
+ "PROPN|NameType=Com|r-obl": 875,
2008
+ "PROPN|NameType=Geo": 876,
2009
+ "PROPN|NameType=Geo|l-nsubj": 877,
2010
+ "PROPN|NameType=Geo|l-obl": 878,
2011
+ "PROPN|NameType=Geo|r-compound": 879,
2012
+ "PROPN|NameType=Geo|r-conj": 880,
2013
+ "PROPN|NameType=Geo|r-flat": 881,
2014
+ "PROPN|NameType=Geo|r-list": 882,
2015
+ "PROPN|NameType=Geo|r-nmod": 883,
2016
+ "PROPN|NameType=Geo|r-nsubj": 884,
2017
+ "PROPN|NameType=Geo|r-nummod": 885,
2018
+ "PROPN|NameType=Geo|r-obj": 886,
2019
+ "PROPN|NameType=Geo|r-obl": 887,
2020
+ "PROPN|NameType=Geo|root": 888,
2021
+ "PROPN|NameType=Giv": 889,
2022
+ "PROPN|NameType=Giv|l-dislocated": 890,
2023
+ "PROPN|NameType=Giv|l-nsubj": 891,
2024
+ "PROPN|NameType=Giv|l-obl": 892,
2025
+ "PROPN|NameType=Giv|r-acl": 893,
2026
+ "PROPN|NameType=Giv|r-appos": 894,
2027
+ "PROPN|NameType=Giv|r-ccomp": 895,
2028
+ "PROPN|NameType=Giv|r-conj": 896,
2029
+ "PROPN|NameType=Giv|r-flat": 897,
2030
+ "PROPN|NameType=Giv|r-list": 898,
2031
+ "PROPN|NameType=Giv|r-nmod": 899,
2032
+ "PROPN|NameType=Giv|r-nsubj": 900,
2033
+ "PROPN|NameType=Giv|r-obj": 901,
2034
+ "PROPN|NameType=Giv|r-obl": 902,
2035
+ "PROPN|NameType=Giv|root": 903,
2036
+ "PROPN|NameType=Nat": 904,
2037
+ "PROPN|NameType=Nat|l-csubj": 905,
2038
+ "PROPN|NameType=Nat|l-nsubj": 906,
2039
+ "PROPN|NameType=Nat|l-obl": 907,
2040
+ "PROPN|NameType=Nat|r-acl": 908,
2041
+ "PROPN|NameType=Nat|r-appos": 909,
2042
+ "PROPN|NameType=Nat|r-compound": 910,
2043
+ "PROPN|NameType=Nat|r-conj": 911,
2044
+ "PROPN|NameType=Nat|r-flat": 912,
2045
+ "PROPN|NameType=Nat|r-list": 913,
2046
+ "PROPN|NameType=Nat|r-nmod": 914,
2047
+ "PROPN|NameType=Nat|r-nummod": 915,
2048
+ "PROPN|NameType=Nat|r-obj": 916,
2049
+ "PROPN|NameType=Nat|r-obl": 917,
2050
+ "PROPN|NameType=Oth": 918,
2051
+ "PROPN|NameType=Oth|l-dislocated": 919,
2052
+ "PROPN|NameType=Oth|l-nsubj": 920,
2053
+ "PROPN|NameType=Oth|r-acl": 921,
2054
+ "PROPN|NameType=Oth|r-appos": 922,
2055
+ "PROPN|NameType=Oth|r-compound": 923,
2056
+ "PROPN|NameType=Oth|r-conj": 924,
2057
+ "PROPN|NameType=Oth|r-flat": 925,
2058
+ "PROPN|NameType=Oth|r-nmod": 926,
2059
+ "PROPN|NameType=Oth|r-obj": 927,
2060
+ "PROPN|NameType=Oth|r-obl": 928,
2061
+ "PROPN|NameType=Oth|root": 929,
2062
+ "PROPN|NameType=Pro": 930,
2063
+ "PROPN|NameType=Pro|l-nsubj": 931,
2064
+ "PROPN|NameType=Pro|l-obl": 932,
2065
+ "PROPN|NameType=Pro|r-advcl": 933,
2066
+ "PROPN|NameType=Pro|r-flat": 934,
2067
+ "PROPN|NameType=Pro|r-nmod": 935,
2068
+ "PROPN|NameType=Pro|r-obj": 936,
2069
+ "PROPN|NameType=Prs": 937,
2070
+ "PROPN|NameType=Prs|l-dislocated": 938,
2071
+ "PROPN|NameType=Prs|l-nsubj": 939,
2072
+ "PROPN|NameType=Prs|l-obl": 940,
2073
+ "PROPN|NameType=Prs|l-vocative": 941,
2074
+ "PROPN|NameType=Prs|r-conj": 942,
2075
+ "PROPN|NameType=Prs|r-discourse": 943,
2076
+ "PROPN|NameType=Prs|r-flat": 944,
2077
+ "PROPN|NameType=Prs|r-list": 945,
2078
+ "PROPN|NameType=Prs|r-nmod": 946,
2079
+ "PROPN|NameType=Prs|r-obj": 947,
2080
+ "PROPN|NameType=Prs|r-obl": 948,
2081
+ "PROPN|NameType=Prs|r-vocative": 949,
2082
+ "PROPN|NameType=Sur": 950,
2083
+ "PROPN|NameType=Sur|l-nsubj": 951,
2084
+ "PROPN|NameType=Sur|r-flat": 952,
2085
+ "PROPN|NameType=Sur|r-nmod": 953,
2086
+ "PROPN|NounType=Class": 954,
2087
+ "PROPN|NounType=Class|r-clf": 955,
2088
+ "PROPN|Prefix=Yes": 956,
2089
+ "PROPN|Prefix=Yes|l-nsubj": 957,
2090
+ "PROPN|Prefix=Yes|r-nmod": 958,
2091
+ "PROPN|l-advmod": 959,
2092
+ "PROPN|l-aux": 960,
2093
+ "PROPN|l-nsubj": 961,
2094
+ "PROPN|l-obl": 962,
2095
+ "PROPN|r-acl": 963,
2096
+ "PROPN|r-advmod": 964,
2097
+ "PROPN|r-appos": 965,
2098
+ "PROPN|r-clf": 966,
2099
+ "PROPN|r-compound": 967,
2100
+ "PROPN|r-conj": 968,
2101
+ "PROPN|r-fixed": 969,
2102
+ "PROPN|r-flat": 970,
2103
+ "PROPN|r-flat:name": 971,
2104
+ "PROPN|r-iobj": 972,
2105
+ "PROPN|r-list": 973,
2106
+ "PROPN|r-nmod": 974,
2107
+ "PROPN|r-nmod:poss": 975,
2108
+ "PROPN|r-obj": 976,
2109
+ "PROPN|r-obl": 977,
2110
+ "PROPN|root": 978,
2111
+ "PUNCT": 979,
2112
+ "PUNCT|NounType=Class": 980,
2113
+ "PUNCT|NounType=Class|r-punct": 981,
2114
+ "PUNCT|l-advmod": 982,
2115
+ "PUNCT|l-dep": 983,
2116
+ "PUNCT|l-punct": 984,
2117
+ "PUNCT|r-advmod": 985,
2118
+ "PUNCT|r-clf": 986,
2119
+ "PUNCT|r-dep": 987,
2120
+ "PUNCT|r-punct": 988,
2121
+ "PUNCT|root": 989,
2122
+ "SCONJ": 990,
2123
+ "SCONJ|NumType=Mult": 991,
2124
+ "SCONJ|NumType=Mult|l-mark": 992,
2125
+ "SCONJ|Prefix=Yes": 993,
2126
+ "SCONJ|Prefix=Yes|l-cc": 994,
2127
+ "SCONJ|Prefix=Yes|l-mark": 995,
2128
+ "SCONJ|VerbType=Cop": 996,
2129
+ "SCONJ|VerbType=Cop|l-mark": 997,
2130
+ "SCONJ|l-advmod": 998,
2131
+ "SCONJ|l-case": 999,
2132
+ "SCONJ|l-cc": 1000,
2133
+ "SCONJ|l-discourse": 1001,
2134
+ "SCONJ|l-mark": 1002,
2135
+ "SCONJ|l-nsubj": 1003,
2136
+ "SCONJ|l-orphan": 1004,
2137
+ "SCONJ|r-advcl": 1005,
2138
+ "SCONJ|r-compound": 1006,
2139
+ "SCONJ|r-fixed": 1007,
2140
+ "SCONJ|r-flat": 1008,
2141
+ "SCONJ|r-mark": 1009,
2142
+ "SCONJ|r-orphan": 1010,
2143
+ "SCONJ|root": 1011,
2144
+ "SYM": 1012,
2145
+ "SYM|l-dep": 1013,
2146
+ "SYM|r-clf": 1014,
2147
+ "SYM|r-nmod": 1015,
2148
+ "SYM|r-obj": 1016,
2149
+ "SYM|r-obl": 1017,
2150
+ "SYM|r-xcomp": 1018,
2151
+ "VERB": 1019,
2152
+ "VERB|Abbr=Yes": 1020,
2153
+ "VERB|Abbr=Yes|r-acl": 1021,
2154
+ "VERB|Foreign=Yes": 1022,
2155
+ "VERB|Foreign=Yes|l-nsubj": 1023,
2156
+ "VERB|Foreign=Yes|r-acl": 1024,
2157
+ "VERB|Foreign=Yes|r-advcl": 1025,
2158
+ "VERB|Foreign=Yes|r-ccomp": 1026,
2159
+ "VERB|Foreign=Yes|r-compound": 1027,
2160
+ "VERB|Foreign=Yes|r-conj": 1028,
2161
+ "VERB|Foreign=Yes|r-flat": 1029,
2162
+ "VERB|Foreign=Yes|r-nmod": 1030,
2163
+ "VERB|Foreign=Yes|r-xcomp": 1031,
2164
+ "VERB|Foreign=Yes|root": 1032,
2165
+ "VERB|Mood=Imp": 1033,
2166
+ "VERB|Mood=Imp|r-xcomp": 1034,
2167
+ "VERB|NounType=Class": 1035,
2168
+ "VERB|NounType=Class|r-acl": 1036,
2169
+ "VERB|NounType=Class|r-compound": 1037,
2170
+ "VERB|PartType=Adj": 1038,
2171
+ "VERB|PartType=Adj|r-acl": 1039,
2172
+ "VERB|Prefix=Yes": 1040,
2173
+ "VERB|Prefix=Yes|l-acl": 1041,
2174
+ "VERB|Prefix=Yes|l-nsubj": 1042,
2175
+ "VERB|Prefix=Yes|r-acl": 1043,
2176
+ "VERB|Prefix=Yes|r-advcl": 1044,
2177
+ "VERB|Prefix=Yes|r-ccomp": 1045,
2178
+ "VERB|Prefix=Yes|r-compound": 1046,
2179
+ "VERB|Prefix=Yes|r-conj": 1047,
2180
+ "VERB|Prefix=Yes|r-parataxis": 1048,
2181
+ "VERB|Prefix=Yes|root": 1049,
2182
+ "VERB|VerbType=Cop": 1050,
2183
+ "VERB|VerbType=Cop|l-advmod": 1051,
2184
+ "VERB|VerbType=Cop|l-cop": 1052,
2185
+ "VERB|VerbType=Cop|r-acl": 1053,
2186
+ "VERB|VerbType=Cop|r-advcl": 1054,
2187
+ "VERB|VerbType=Cop|r-ccomp": 1055,
2188
+ "VERB|VerbType=Cop|r-compound": 1056,
2189
+ "VERB|VerbType=Cop|r-parataxis": 1057,
2190
+ "VERB|VerbType=Cop|root": 1058,
2191
+ "VERB|Voice=Pass": 1059,
2192
+ "VERB|Voice=Pass|l-aux:pass": 1060,
2193
+ "VERB|l-acl": 1061,
2194
+ "VERB|l-advcl": 1062,
2195
+ "VERB|l-advmod": 1063,
2196
+ "VERB|l-aux": 1064,
2197
+ "VERB|l-case": 1065,
2198
+ "VERB|l-cc": 1066,
2199
+ "VERB|l-ccomp": 1067,
2200
+ "VERB|l-compound": 1068,
2201
+ "VERB|l-conj": 1069,
2202
+ "VERB|l-cop": 1070,
2203
+ "VERB|l-csubj": 1071,
2204
+ "VERB|l-discourse": 1072,
2205
+ "VERB|l-dislocated": 1073,
2206
+ "VERB|l-mark": 1074,
2207
+ "VERB|l-nsubj": 1075,
2208
+ "VERB|l-obl": 1076,
2209
+ "VERB|l-orphan": 1077,
2210
+ "VERB|l-xcomp": 1078,
2211
+ "VERB|r-acl": 1079,
2212
+ "VERB|r-acl:relcl": 1080,
2213
+ "VERB|r-advcl": 1081,
2214
+ "VERB|r-advmod": 1082,
2215
+ "VERB|r-appos": 1083,
2216
+ "VERB|r-aux": 1084,
2217
+ "VERB|r-case": 1085,
2218
+ "VERB|r-cc": 1086,
2219
+ "VERB|r-ccomp": 1087,
2220
+ "VERB|r-clf": 1088,
2221
+ "VERB|r-compound": 1089,
2222
+ "VERB|r-conj": 1090,
2223
+ "VERB|r-dep": 1091,
2224
+ "VERB|r-det": 1092,
2225
+ "VERB|r-discourse": 1093,
2226
+ "VERB|r-fixed": 1094,
2227
+ "VERB|r-flat": 1095,
2228
+ "VERB|r-list": 1096,
2229
+ "VERB|r-mark": 1097,
2230
+ "VERB|r-nmod": 1098,
2231
+ "VERB|r-nsubj": 1099,
2232
+ "VERB|r-obj": 1100,
2233
+ "VERB|r-obl": 1101,
2234
+ "VERB|r-orphan": 1102,
2235
+ "VERB|r-parataxis": 1103,
2236
+ "VERB|r-punct": 1104,
2237
+ "VERB|r-xcomp": 1105,
2238
+ "VERB|root": 1106,
2239
+ "X": 1107
2240
+ },
2241
+ "layer_norm_epsilon": 1e-05,
2242
+ "model_type": "gpt2",
2243
+ "n_ctx": 512,
2244
+ "n_embd": 768,
2245
+ "n_head": 12,
2246
+ "n_inner": 3072,
2247
+ "n_layer": 12,
2248
+ "n_positions": 512,
2249
+ "pad_token_id": 50002,
2250
+ "prefix": "[CLS]",
2251
+ "reorder_and_upcast_attn": false,
2252
+ "resid_pdrop": 0.1,
2253
+ "scale_attn_by_inverse_layer_idx": false,
2254
+ "scale_attn_weights": true,
2255
+ "summary_activation": null,
2256
+ "summary_first_dropout": 0.1,
2257
+ "summary_proj_to_labels": true,
2258
+ "summary_type": "cls_index",
2259
+ "summary_use_proj": true,
2260
+ "tokenizer_class": "PreTrainedTokenizerFast",
2261
+ "torch_dtype": "float32",
2262
+ "transformers_version": "4.44.0",
2263
+ "use_cache": true,
2264
+ "vocab_size": 51200
2265
+ }
maker.py ADDED
@@ -0,0 +1,138 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #! /usr/bin/python3
2
+ src="goldfish-models/tha_thai_1000mb"
3
+ tgt="KoichiYasuoka/goldfish-gpt2-thai-ud-causal"
4
+ url="https://github.com/KoichiYasuoka/spaCy-Thai"
5
+
6
+ import os,json,re
7
+ from transformers import AutoTokenizer,PreTrainedTokenizerFast
8
+ from tokenizers import pre_tokenizers,decoders
9
+ d=os.path.join(os.path.basename(url),"UD_Thai-Corpora")
10
+ os.system("test -d "+d+" || git clone --depth=1 "+url)
11
+ os.system("for F in train dev test ; do cat "+d+"/*-$F*.conllu > $F.conllu ; done")
12
+ tkz=AutoTokenizer.from_pretrained(src,add_prefix_space=False,legacy=False,model_max_length=768)
13
+ tkz.backend_tokenizer.pre_tokenizer=pre_tokenizers.Metaspace(prepend_scheme="never")
14
+ tkz.backend_tokenizer.decoder=decoders.Metaspace(prepend_scheme="never")
15
+ tkz.save_pretrained("tmpdir")
16
+ d=json.loads(tkz.backend_tokenizer.to_str())
17
+ form=set()
18
+ for t in tkz.special_tokens_map.values():
19
+ if type(t)==list:
20
+ for k in t:
21
+ form.add(k)
22
+ else:
23
+ form.add(t)
24
+ with open("train.conllu","r",encoding="utf-8") as r:
25
+ for s in r:
26
+ w=s.split("\t")
27
+ if len(w)==10 and w[0].isdecimal():
28
+ form.add(w[1])
29
+ tcc=re.compile("^[\u0e40-\u0e44]?[\u0e01-\u0e2e][\u0e30-\u0e3a\u0e45\u0e47-\u0e4e]*$")
30
+ for t in d["model"]["vocab"]:
31
+ if len(t[0])>1 and t[0] not in form:
32
+ if not tcc.match(t[0]):
33
+ t[1]*=len(t[0])
34
+ tkz.backend_tokenizer.from_str(json.dumps(d)).save("tmpdir/tokenizer.json")
35
+ tkz=PreTrainedTokenizerFast.from_pretrained("tmpdir")
36
+
37
+ class UDCausalDataset(object):
38
+ def __init__(self,conllu,tokenizer,embeddings=None):
39
+ self.conllu=open(conllu,"r",encoding="utf-8")
40
+ self.tokenizer=tokenizer
41
+ self.embeddings=embeddings
42
+ self.max_tokens=3
43
+ self.seeks=[(0,0)]
44
+ label=set(["SYM"])
45
+ dep=set()
46
+ s=self.conllu.readline()
47
+ while s!="":
48
+ if s=="\n":
49
+ self.seeks.append((self.conllu.tell(),0))
50
+ else:
51
+ w=s.split("\t")
52
+ if len(w)==10:
53
+ if w[0].isdecimal():
54
+ p=w[3] if w[5]=="_" else w[3]+"|"+w[5]
55
+ label.add(p)
56
+ if w[6].isdecimal():
57
+ dep.add(p+("|" if w[6]=="0" else "|l-" if int(w[0])<int(w[6]) else "|r-")+w[7])
58
+ self.seeks.append((self.seeks[-1][0],int(w[0])))
59
+ self.max_tokens=max(self.max_tokens,int(w[0])*2+1)
60
+ s=self.conllu.readline()
61
+ lid={}
62
+ for i,l in enumerate(sorted(label)):
63
+ lid[l],lid["B-"+l],lid["I-"+l]=i*3,i*3+1,i*3+2
64
+ for i,d in enumerate(sorted(dep),len(lid)):
65
+ lid[d]=i
66
+ self.label2id=lid
67
+ def __call__(*args):
68
+ lid={l:i for i,l in enumerate(sorted(set(sum([list(t.label2id) for t in args],[]))))}
69
+ for t in args:
70
+ t.label2id=lid
71
+ return lid
72
+ def __del__(self):
73
+ self.conllu.close()
74
+ __len__=lambda self:len(self.seeks)-1
75
+ def __getitem__(self,i):
76
+ s,t=self.seeks[i]
77
+ self.conllu.seek(s)
78
+ form,upos,deps,w=[],[],[],[""]
79
+ while w[0]!="\n":
80
+ w=self.conllu.readline().split("\t")
81
+ if len(w)==10:
82
+ form.append(w[1])
83
+ if w[0].isdecimal():
84
+ upos.append(w[3] if w[5]=="_" else w[3]+"|"+w[5])
85
+ if t>0:
86
+ deps.append((int(w[6]),w[7]))
87
+ v=self.tokenizer(form,add_special_tokens=False)
88
+ if t==0:
89
+ i,u=[self.tokenizer.cls_token_id],["SYM"]
90
+ for j,(x,y) in enumerate(zip(v["input_ids"],upos)):
91
+ if x!=[]:
92
+ i+=x
93
+ u+=[y] if len(x)==1 else ["B-"+y]+["I-"+y]*(len(x)-1)
94
+ emb=self.embeddings
95
+ pad=self.tokenizer.pad_token_id
96
+ else:
97
+ import torch
98
+ m=[]
99
+ for x in v["input_ids"]:
100
+ if x==[]:
101
+ m.append(self.embeddings[self.tokenizer.unk_token_id,:])
102
+ else:
103
+ m.append(self.embeddings[x,:].sum(axis=0))
104
+ m.append(self.embeddings[self.tokenizer.sep_token_id,:])
105
+ m.append(self.embeddings[self.tokenizer.pad_token_id,:])
106
+ m.append(self.embeddings[self.tokenizer.cls_token_id,:])
107
+ emb=torch.stack(m)
108
+ i,u=list(range(-1,len(upos)+1)),["SYM"]+upos+["SYM"]
109
+ i.append(t-1)
110
+ k,d=deps[t-1]
111
+ u.append(upos[t-1]+"|"+d if k==0 else upos[t-1])
112
+ for j in range(t,len(upos)):
113
+ i.append(j)
114
+ a,b=deps[j]
115
+ u.append(upos[j]+"|r-"+b if a==t else upos[t-1]+"|l-"+d if j+1==k else upos[j])
116
+ pad=-2
117
+ j=self.max_tokens-len(i)
118
+ if j>0:
119
+ ids=i+[pad]*j
120
+ upos=u+["SYM"]*j
121
+ else:
122
+ ids=i[0:self.max_tokens]
123
+ upos=u[0:self.max_tokens]
124
+ return {"inputs_embeds":emb[ids,:],"labels":[self.label2id[p] for p in upos]}
125
+
126
+ trainDS=UDCausalDataset("train.conllu",tkz)
127
+ devDS=UDCausalDataset("dev.conllu",tkz)
128
+ testDS=UDCausalDataset("test.conllu",tkz)
129
+ lid=trainDS(devDS,testDS)
130
+ cfg=AutoConfig.from_pretrained(src,num_labels=len(lid),label2id=lid,id2label={i:l for l,i in lid.items()},ignore_mismatched_sizes=True)
131
+ mdl=GPT2ForTokenClassification.from_pretrained(src,config=cfg,ignore_mismatched_sizes=True)
132
+ trainDS.embeddings=mdl.get_input_embeddings().weight
133
+ trainDS.max_tokens=min(trainDS.max_tokens,cfg.max_position_embeddings)
134
+ arg=TrainingArguments(num_train_epochs=3,per_device_train_batch_size=32,dataloader_pin_memory=False,output_dir=tgt,overwrite_output_dir=True,save_total_limit=2,learning_rate=5e-05,warmup_ratio=0.1,save_safetensors=False)
135
+ trn=Trainer(args=arg,data_collator=DefaultDataCollator(),model=mdl,train_dataset=trainDS)
136
+ trn.train()
137
+ trn.save_model(tgt)
138
+ tkz.save_pretrained(tgt)
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b9e4284b422aa0f64d430317a7d9b642bc7b0f4bd72d2a988c2ef5c9a46d633
3
+ size 502540386
special_tokens_map.json ADDED
@@ -0,0 +1,1249 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "[XXXXX0]",
4
+ "[XXXXX1]",
5
+ "[XXXXX2]",
6
+ "[XXXXX3]",
7
+ "[XXXXX4]",
8
+ "[XXXXX5]",
9
+ "[XXXXX6]",
10
+ "[XXXXX7]",
11
+ "[XXXXX8]",
12
+ "[XXXXX9]",
13
+ "[XXXXX10]",
14
+ "[XXXXX11]",
15
+ "[XXXXX12]",
16
+ "[XXXXX13]",
17
+ "[XXXXX14]",
18
+ "[XXXXX15]",
19
+ "[XXXXX16]",
20
+ "[XXXXX17]",
21
+ "[XXXXX18]",
22
+ "[XXXXX19]",
23
+ "[XXXXX20]",
24
+ "[XXXXX21]",
25
+ "[XXXXX22]",
26
+ "[XXXXX23]",
27
+ "[XXXXX24]",
28
+ "[XXXXX25]",
29
+ "[XXXXX26]",
30
+ "[XXXXX27]",
31
+ "[XXXXX28]",
32
+ "[XXXXX29]",
33
+ "[XXXXX30]",
34
+ "[XXXXX31]",
35
+ "[XXXXX32]",
36
+ "[XXXXX33]",
37
+ "[XXXXX34]",
38
+ "[XXXXX35]",
39
+ "[XXXXX36]",
40
+ "[XXXXX37]",
41
+ "[XXXXX38]",
42
+ "[XXXXX39]",
43
+ "[XXXXX40]",
44
+ "[XXXXX41]",
45
+ "[XXXXX42]",
46
+ "[XXXXX43]",
47
+ "[XXXXX44]",
48
+ "[XXXXX45]",
49
+ "[XXXXX46]",
50
+ "[XXXXX47]",
51
+ "[XXXXX48]",
52
+ "[XXXXX49]",
53
+ "[XXXXX50]",
54
+ "[XXXXX51]",
55
+ "[XXXXX52]",
56
+ "[XXXXX53]",
57
+ "[XXXXX54]",
58
+ "[XXXXX55]",
59
+ "[XXXXX56]",
60
+ "[XXXXX57]",
61
+ "[XXXXX58]",
62
+ "[XXXXX59]",
63
+ "[XXXXX60]",
64
+ "[XXXXX61]",
65
+ "[XXXXX62]",
66
+ "[XXXXX63]",
67
+ "[XXXXX64]",
68
+ "[XXXXX65]",
69
+ "[XXXXX66]",
70
+ "[XXXXX67]",
71
+ "[XXXXX68]",
72
+ "[XXXXX69]",
73
+ "[XXXXX70]",
74
+ "[XXXXX71]",
75
+ "[XXXXX72]",
76
+ "[XXXXX73]",
77
+ "[XXXXX74]",
78
+ "[XXXXX75]",
79
+ "[XXXXX76]",
80
+ "[XXXXX77]",
81
+ "[XXXXX78]",
82
+ "[XXXXX79]",
83
+ "[XXXXX80]",
84
+ "[XXXXX81]",
85
+ "[XXXXX82]",
86
+ "[XXXXX83]",
87
+ "[XXXXX84]",
88
+ "[XXXXX85]",
89
+ "[XXXXX86]",
90
+ "[XXXXX87]",
91
+ "[XXXXX88]",
92
+ "[XXXXX89]",
93
+ "[XXXXX90]",
94
+ "[XXXXX91]",
95
+ "[XXXXX92]",
96
+ "[XXXXX93]",
97
+ "[XXXXX94]",
98
+ "[XXXXX95]",
99
+ "[XXXXX96]",
100
+ "[XXXXX97]",
101
+ "[XXXXX98]",
102
+ "[XXXXX99]",
103
+ "[XXXXX100]",
104
+ "[XXXXX101]",
105
+ "[XXXXX102]",
106
+ "[XXXXX103]",
107
+ "[XXXXX104]",
108
+ "[XXXXX105]",
109
+ "[XXXXX106]",
110
+ "[XXXXX107]",
111
+ "[XXXXX108]",
112
+ "[XXXXX109]",
113
+ "[XXXXX110]",
114
+ "[XXXXX111]",
115
+ "[XXXXX112]",
116
+ "[XXXXX113]",
117
+ "[XXXXX114]",
118
+ "[XXXXX115]",
119
+ "[XXXXX116]",
120
+ "[XXXXX117]",
121
+ "[XXXXX118]",
122
+ "[XXXXX119]",
123
+ "[XXXXX120]",
124
+ "[XXXXX121]",
125
+ "[XXXXX122]",
126
+ "[XXXXX123]",
127
+ "[XXXXX124]",
128
+ "[XXXXX125]",
129
+ "[XXXXX126]",
130
+ "[XXXXX127]",
131
+ "[XXXXX128]",
132
+ "[XXXXX129]",
133
+ "[XXXXX130]",
134
+ "[XXXXX131]",
135
+ "[XXXXX132]",
136
+ "[XXXXX133]",
137
+ "[XXXXX134]",
138
+ "[XXXXX135]",
139
+ "[XXXXX136]",
140
+ "[XXXXX137]",
141
+ "[XXXXX138]",
142
+ "[XXXXX139]",
143
+ "[XXXXX140]",
144
+ "[XXXXX141]",
145
+ "[XXXXX142]",
146
+ "[XXXXX143]",
147
+ "[XXXXX144]",
148
+ "[XXXXX145]",
149
+ "[XXXXX146]",
150
+ "[XXXXX147]",
151
+ "[XXXXX148]",
152
+ "[XXXXX149]",
153
+ "[XXXXX150]",
154
+ "[XXXXX151]",
155
+ "[XXXXX152]",
156
+ "[XXXXX153]",
157
+ "[XXXXX154]",
158
+ "[XXXXX155]",
159
+ "[XXXXX156]",
160
+ "[XXXXX157]",
161
+ "[XXXXX158]",
162
+ "[XXXXX159]",
163
+ "[XXXXX160]",
164
+ "[XXXXX161]",
165
+ "[XXXXX162]",
166
+ "[XXXXX163]",
167
+ "[XXXXX164]",
168
+ "[XXXXX165]",
169
+ "[XXXXX166]",
170
+ "[XXXXX167]",
171
+ "[XXXXX168]",
172
+ "[XXXXX169]",
173
+ "[XXXXX170]",
174
+ "[XXXXX171]",
175
+ "[XXXXX172]",
176
+ "[XXXXX173]",
177
+ "[XXXXX174]",
178
+ "[XXXXX175]",
179
+ "[XXXXX176]",
180
+ "[XXXXX177]",
181
+ "[XXXXX178]",
182
+ "[XXXXX179]",
183
+ "[XXXXX180]",
184
+ "[XXXXX181]",
185
+ "[XXXXX182]",
186
+ "[XXXXX183]",
187
+ "[XXXXX184]",
188
+ "[XXXXX185]",
189
+ "[XXXXX186]",
190
+ "[XXXXX187]",
191
+ "[XXXXX188]",
192
+ "[XXXXX189]",
193
+ "[XXXXX190]",
194
+ "[XXXXX191]",
195
+ "[XXXXX192]",
196
+ "[XXXXX193]",
197
+ "[XXXXX194]",
198
+ "[XXXXX195]",
199
+ "[XXXXX196]",
200
+ "[XXXXX197]",
201
+ "[XXXXX198]",
202
+ "[XXXXX199]",
203
+ "[XXXXX200]",
204
+ "[XXXXX201]",
205
+ "[XXXXX202]",
206
+ "[XXXXX203]",
207
+ "[XXXXX204]",
208
+ "[XXXXX205]",
209
+ "[XXXXX206]",
210
+ "[XXXXX207]",
211
+ "[XXXXX208]",
212
+ "[XXXXX209]",
213
+ "[XXXXX210]",
214
+ "[XXXXX211]",
215
+ "[XXXXX212]",
216
+ "[XXXXX213]",
217
+ "[XXXXX214]",
218
+ "[XXXXX215]",
219
+ "[XXXXX216]",
220
+ "[XXXXX217]",
221
+ "[XXXXX218]",
222
+ "[XXXXX219]",
223
+ "[XXXXX220]",
224
+ "[XXXXX221]",
225
+ "[XXXXX222]",
226
+ "[XXXXX223]",
227
+ "[XXXXX224]",
228
+ "[XXXXX225]",
229
+ "[XXXXX226]",
230
+ "[XXXXX227]",
231
+ "[XXXXX228]",
232
+ "[XXXXX229]",
233
+ "[XXXXX230]",
234
+ "[XXXXX231]",
235
+ "[XXXXX232]",
236
+ "[XXXXX233]",
237
+ "[XXXXX234]",
238
+ "[XXXXX235]",
239
+ "[XXXXX236]",
240
+ "[XXXXX237]",
241
+ "[XXXXX238]",
242
+ "[XXXXX239]",
243
+ "[XXXXX240]",
244
+ "[XXXXX241]",
245
+ "[XXXXX242]",
246
+ "[XXXXX243]",
247
+ "[XXXXX244]",
248
+ "[XXXXX245]",
249
+ "[XXXXX246]",
250
+ "[XXXXX247]",
251
+ "[XXXXX248]",
252
+ "[XXXXX249]",
253
+ "[XXXXX250]",
254
+ "[XXXXX251]",
255
+ "[XXXXX252]",
256
+ "[XXXXX253]",
257
+ "[XXXXX254]",
258
+ "[XXXXX255]",
259
+ "[XXXXX256]",
260
+ "[XXXXX257]",
261
+ "[XXXXX258]",
262
+ "[XXXXX259]",
263
+ "[XXXXX260]",
264
+ "[XXXXX261]",
265
+ "[XXXXX262]",
266
+ "[XXXXX263]",
267
+ "[XXXXX264]",
268
+ "[XXXXX265]",
269
+ "[XXXXX266]",
270
+ "[XXXXX267]",
271
+ "[XXXXX268]",
272
+ "[XXXXX269]",
273
+ "[XXXXX270]",
274
+ "[XXXXX271]",
275
+ "[XXXXX272]",
276
+ "[XXXXX273]",
277
+ "[XXXXX274]",
278
+ "[XXXXX275]",
279
+ "[XXXXX276]",
280
+ "[XXXXX277]",
281
+ "[XXXXX278]",
282
+ "[XXXXX279]",
283
+ "[XXXXX280]",
284
+ "[XXXXX281]",
285
+ "[XXXXX282]",
286
+ "[XXXXX283]",
287
+ "[XXXXX284]",
288
+ "[XXXXX285]",
289
+ "[XXXXX286]",
290
+ "[XXXXX287]",
291
+ "[XXXXX288]",
292
+ "[XXXXX289]",
293
+ "[XXXXX290]",
294
+ "[XXXXX291]",
295
+ "[XXXXX292]",
296
+ "[XXXXX293]",
297
+ "[XXXXX294]",
298
+ "[XXXXX295]",
299
+ "[XXXXX296]",
300
+ "[XXXXX297]",
301
+ "[XXXXX298]",
302
+ "[XXXXX299]",
303
+ "[XXXXX300]",
304
+ "[XXXXX301]",
305
+ "[XXXXX302]",
306
+ "[XXXXX303]",
307
+ "[XXXXX304]",
308
+ "[XXXXX305]",
309
+ "[XXXXX306]",
310
+ "[XXXXX307]",
311
+ "[XXXXX308]",
312
+ "[XXXXX309]",
313
+ "[XXXXX310]",
314
+ "[XXXXX311]",
315
+ "[XXXXX312]",
316
+ "[XXXXX313]",
317
+ "[XXXXX314]",
318
+ "[XXXXX315]",
319
+ "[XXXXX316]",
320
+ "[XXXXX317]",
321
+ "[XXXXX318]",
322
+ "[XXXXX319]",
323
+ "[XXXXX320]",
324
+ "[XXXXX321]",
325
+ "[XXXXX322]",
326
+ "[XXXXX323]",
327
+ "[XXXXX324]",
328
+ "[XXXXX325]",
329
+ "[XXXXX326]",
330
+ "[XXXXX327]",
331
+ "[XXXXX328]",
332
+ "[XXXXX329]",
333
+ "[XXXXX330]",
334
+ "[XXXXX331]",
335
+ "[XXXXX332]",
336
+ "[XXXXX333]",
337
+ "[XXXXX334]",
338
+ "[XXXXX335]",
339
+ "[XXXXX336]",
340
+ "[XXXXX337]",
341
+ "[XXXXX338]",
342
+ "[XXXXX339]",
343
+ "[XXXXX340]",
344
+ "[XXXXX341]",
345
+ "[XXXXX342]",
346
+ "[XXXXX343]",
347
+ "[XXXXX344]",
348
+ "[XXXXX345]",
349
+ "[XXXXX346]",
350
+ "[XXXXX347]",
351
+ "[XXXXX348]",
352
+ "[XXXXX349]",
353
+ "[XXXXX350]",
354
+ "[XXXXX351]",
355
+ "[XXXXX352]",
356
+ "[XXXXX353]",
357
+ "[XXXXX354]",
358
+ "[XXXXX355]",
359
+ "[XXXXX356]",
360
+ "[XXXXX357]",
361
+ "[XXXXX358]",
362
+ "[XXXXX359]",
363
+ "[XXXXX360]",
364
+ "[XXXXX361]",
365
+ "[XXXXX362]",
366
+ "[XXXXX363]",
367
+ "[XXXXX364]",
368
+ "[XXXXX365]",
369
+ "[XXXXX366]",
370
+ "[XXXXX367]",
371
+ "[XXXXX368]",
372
+ "[XXXXX369]",
373
+ "[XXXXX370]",
374
+ "[XXXXX371]",
375
+ "[XXXXX372]",
376
+ "[XXXXX373]",
377
+ "[XXXXX374]",
378
+ "[XXXXX375]",
379
+ "[XXXXX376]",
380
+ "[XXXXX377]",
381
+ "[XXXXX378]",
382
+ "[XXXXX379]",
383
+ "[XXXXX380]",
384
+ "[XXXXX381]",
385
+ "[XXXXX382]",
386
+ "[XXXXX383]",
387
+ "[XXXXX384]",
388
+ "[XXXXX385]",
389
+ "[XXXXX386]",
390
+ "[XXXXX387]",
391
+ "[XXXXX388]",
392
+ "[XXXXX389]",
393
+ "[XXXXX390]",
394
+ "[XXXXX391]",
395
+ "[XXXXX392]",
396
+ "[XXXXX393]",
397
+ "[XXXXX394]",
398
+ "[XXXXX395]",
399
+ "[XXXXX396]",
400
+ "[XXXXX397]",
401
+ "[XXXXX398]",
402
+ "[XXXXX399]",
403
+ "[XXXXX400]",
404
+ "[XXXXX401]",
405
+ "[XXXXX402]",
406
+ "[XXXXX403]",
407
+ "[XXXXX404]",
408
+ "[XXXXX405]",
409
+ "[XXXXX406]",
410
+ "[XXXXX407]",
411
+ "[XXXXX408]",
412
+ "[XXXXX409]",
413
+ "[XXXXX410]",
414
+ "[XXXXX411]",
415
+ "[XXXXX412]",
416
+ "[XXXXX413]",
417
+ "[XXXXX414]",
418
+ "[XXXXX415]",
419
+ "[XXXXX416]",
420
+ "[XXXXX417]",
421
+ "[XXXXX418]",
422
+ "[XXXXX419]",
423
+ "[XXXXX420]",
424
+ "[XXXXX421]",
425
+ "[XXXXX422]",
426
+ "[XXXXX423]",
427
+ "[XXXXX424]",
428
+ "[XXXXX425]",
429
+ "[XXXXX426]",
430
+ "[XXXXX427]",
431
+ "[XXXXX428]",
432
+ "[XXXXX429]",
433
+ "[XXXXX430]",
434
+ "[XXXXX431]",
435
+ "[XXXXX432]",
436
+ "[XXXXX433]",
437
+ "[XXXXX434]",
438
+ "[XXXXX435]",
439
+ "[XXXXX436]",
440
+ "[XXXXX437]",
441
+ "[XXXXX438]",
442
+ "[XXXXX439]",
443
+ "[XXXXX440]",
444
+ "[XXXXX441]",
445
+ "[XXXXX442]",
446
+ "[XXXXX443]",
447
+ "[XXXXX444]",
448
+ "[XXXXX445]",
449
+ "[XXXXX446]",
450
+ "[XXXXX447]",
451
+ "[XXXXX448]",
452
+ "[XXXXX449]",
453
+ "[XXXXX450]",
454
+ "[XXXXX451]",
455
+ "[XXXXX452]",
456
+ "[XXXXX453]",
457
+ "[XXXXX454]",
458
+ "[XXXXX455]",
459
+ "[XXXXX456]",
460
+ "[XXXXX457]",
461
+ "[XXXXX458]",
462
+ "[XXXXX459]",
463
+ "[XXXXX460]",
464
+ "[XXXXX461]",
465
+ "[XXXXX462]",
466
+ "[XXXXX463]",
467
+ "[XXXXX464]",
468
+ "[XXXXX465]",
469
+ "[XXXXX466]",
470
+ "[XXXXX467]",
471
+ "[XXXXX468]",
472
+ "[XXXXX469]",
473
+ "[XXXXX470]",
474
+ "[XXXXX471]",
475
+ "[XXXXX472]",
476
+ "[XXXXX473]",
477
+ "[XXXXX474]",
478
+ "[XXXXX475]",
479
+ "[XXXXX476]",
480
+ "[XXXXX477]",
481
+ "[XXXXX478]",
482
+ "[XXXXX479]",
483
+ "[XXXXX480]",
484
+ "[XXXXX481]",
485
+ "[XXXXX482]",
486
+ "[XXXXX483]",
487
+ "[XXXXX484]",
488
+ "[XXXXX485]",
489
+ "[XXXXX486]",
490
+ "[XXXXX487]",
491
+ "[XXXXX488]",
492
+ "[XXXXX489]",
493
+ "[XXXXX490]",
494
+ "[XXXXX491]",
495
+ "[XXXXX492]",
496
+ "[XXXXX493]",
497
+ "[XXXXX494]",
498
+ "[XXXXX495]",
499
+ "[XXXXX496]",
500
+ "[XXXXX497]",
501
+ "[XXXXX498]",
502
+ "[XXXXX499]",
503
+ "[XXXXX500]",
504
+ "[XXXXX501]",
505
+ "[XXXXX502]",
506
+ "[XXXXX503]",
507
+ "[XXXXX504]",
508
+ "[XXXXX505]",
509
+ "[XXXXX506]",
510
+ "[XXXXX507]",
511
+ "[XXXXX508]",
512
+ "[XXXXX509]",
513
+ "[XXXXX510]",
514
+ "[XXXXX511]",
515
+ "[XXXXX512]",
516
+ "[XXXXX513]",
517
+ "[XXXXX514]",
518
+ "[XXXXX515]",
519
+ "[XXXXX516]",
520
+ "[XXXXX517]",
521
+ "[XXXXX518]",
522
+ "[XXXXX519]",
523
+ "[XXXXX520]",
524
+ "[XXXXX521]",
525
+ "[XXXXX522]",
526
+ "[XXXXX523]",
527
+ "[XXXXX524]",
528
+ "[XXXXX525]",
529
+ "[XXXXX526]",
530
+ "[XXXXX527]",
531
+ "[XXXXX528]",
532
+ "[XXXXX529]",
533
+ "[XXXXX530]",
534
+ "[XXXXX531]",
535
+ "[XXXXX532]",
536
+ "[XXXXX533]",
537
+ "[XXXXX534]",
538
+ "[XXXXX535]",
539
+ "[XXXXX536]",
540
+ "[XXXXX537]",
541
+ "[XXXXX538]",
542
+ "[XXXXX539]",
543
+ "[XXXXX540]",
544
+ "[XXXXX541]",
545
+ "[XXXXX542]",
546
+ "[XXXXX543]",
547
+ "[XXXXX544]",
548
+ "[XXXXX545]",
549
+ "[XXXXX546]",
550
+ "[XXXXX547]",
551
+ "[XXXXX548]",
552
+ "[XXXXX549]",
553
+ "[XXXXX550]",
554
+ "[XXXXX551]",
555
+ "[XXXXX552]",
556
+ "[XXXXX553]",
557
+ "[XXXXX554]",
558
+ "[XXXXX555]",
559
+ "[XXXXX556]",
560
+ "[XXXXX557]",
561
+ "[XXXXX558]",
562
+ "[XXXXX559]",
563
+ "[XXXXX560]",
564
+ "[XXXXX561]",
565
+ "[XXXXX562]",
566
+ "[XXXXX563]",
567
+ "[XXXXX564]",
568
+ "[XXXXX565]",
569
+ "[XXXXX566]",
570
+ "[XXXXX567]",
571
+ "[XXXXX568]",
572
+ "[XXXXX569]",
573
+ "[XXXXX570]",
574
+ "[XXXXX571]",
575
+ "[XXXXX572]",
576
+ "[XXXXX573]",
577
+ "[XXXXX574]",
578
+ "[XXXXX575]",
579
+ "[XXXXX576]",
580
+ "[XXXXX577]",
581
+ "[XXXXX578]",
582
+ "[XXXXX579]",
583
+ "[XXXXX580]",
584
+ "[XXXXX581]",
585
+ "[XXXXX582]",
586
+ "[XXXXX583]",
587
+ "[XXXXX584]",
588
+ "[XXXXX585]",
589
+ "[XXXXX586]",
590
+ "[XXXXX587]",
591
+ "[XXXXX588]",
592
+ "[XXXXX589]",
593
+ "[XXXXX590]",
594
+ "[XXXXX591]",
595
+ "[XXXXX592]",
596
+ "[XXXXX593]",
597
+ "[XXXXX594]",
598
+ "[XXXXX595]",
599
+ "[XXXXX596]",
600
+ "[XXXXX597]",
601
+ "[XXXXX598]",
602
+ "[XXXXX599]",
603
+ "[XXXXX600]",
604
+ "[XXXXX601]",
605
+ "[XXXXX602]",
606
+ "[XXXXX603]",
607
+ "[XXXXX604]",
608
+ "[XXXXX605]",
609
+ "[XXXXX606]",
610
+ "[XXXXX607]",
611
+ "[XXXXX608]",
612
+ "[XXXXX609]",
613
+ "[XXXXX610]",
614
+ "[XXXXX611]",
615
+ "[XXXXX612]",
616
+ "[XXXXX613]",
617
+ "[XXXXX614]",
618
+ "[XXXXX615]",
619
+ "[XXXXX616]",
620
+ "[XXXXX617]",
621
+ "[XXXXX618]",
622
+ "[XXXXX619]",
623
+ "[XXXXX620]",
624
+ "[XXXXX621]",
625
+ "[XXXXX622]",
626
+ "[XXXXX623]",
627
+ "[XXXXX624]",
628
+ "[XXXXX625]",
629
+ "[XXXXX626]",
630
+ "[XXXXX627]",
631
+ "[XXXXX628]",
632
+ "[XXXXX629]",
633
+ "[XXXXX630]",
634
+ "[XXXXX631]",
635
+ "[XXXXX632]",
636
+ "[XXXXX633]",
637
+ "[XXXXX634]",
638
+ "[XXXXX635]",
639
+ "[XXXXX636]",
640
+ "[XXXXX637]",
641
+ "[XXXXX638]",
642
+ "[XXXXX639]",
643
+ "[XXXXX640]",
644
+ "[XXXXX641]",
645
+ "[XXXXX642]",
646
+ "[XXXXX643]",
647
+ "[XXXXX644]",
648
+ "[XXXXX645]",
649
+ "[XXXXX646]",
650
+ "[XXXXX647]",
651
+ "[XXXXX648]",
652
+ "[XXXXX649]",
653
+ "[XXXXX650]",
654
+ "[XXXXX651]",
655
+ "[XXXXX652]",
656
+ "[XXXXX653]",
657
+ "[XXXXX654]",
658
+ "[XXXXX655]",
659
+ "[XXXXX656]",
660
+ "[XXXXX657]",
661
+ "[XXXXX658]",
662
+ "[XXXXX659]",
663
+ "[XXXXX660]",
664
+ "[XXXXX661]",
665
+ "[XXXXX662]",
666
+ "[XXXXX663]",
667
+ "[XXXXX664]",
668
+ "[XXXXX665]",
669
+ "[XXXXX666]",
670
+ "[XXXXX667]",
671
+ "[XXXXX668]",
672
+ "[XXXXX669]",
673
+ "[XXXXX670]",
674
+ "[XXXXX671]",
675
+ "[XXXXX672]",
676
+ "[XXXXX673]",
677
+ "[XXXXX674]",
678
+ "[XXXXX675]",
679
+ "[XXXXX676]",
680
+ "[XXXXX677]",
681
+ "[XXXXX678]",
682
+ "[XXXXX679]",
683
+ "[XXXXX680]",
684
+ "[XXXXX681]",
685
+ "[XXXXX682]",
686
+ "[XXXXX683]",
687
+ "[XXXXX684]",
688
+ "[XXXXX685]",
689
+ "[XXXXX686]",
690
+ "[XXXXX687]",
691
+ "[XXXXX688]",
692
+ "[XXXXX689]",
693
+ "[XXXXX690]",
694
+ "[XXXXX691]",
695
+ "[XXXXX692]",
696
+ "[XXXXX693]",
697
+ "[XXXXX694]",
698
+ "[XXXXX695]",
699
+ "[XXXXX696]",
700
+ "[XXXXX697]",
701
+ "[XXXXX698]",
702
+ "[XXXXX699]",
703
+ "[XXXXX700]",
704
+ "[XXXXX701]",
705
+ "[XXXXX702]",
706
+ "[XXXXX703]",
707
+ "[XXXXX704]",
708
+ "[XXXXX705]",
709
+ "[XXXXX706]",
710
+ "[XXXXX707]",
711
+ "[XXXXX708]",
712
+ "[XXXXX709]",
713
+ "[XXXXX710]",
714
+ "[XXXXX711]",
715
+ "[XXXXX712]",
716
+ "[XXXXX713]",
717
+ "[XXXXX714]",
718
+ "[XXXXX715]",
719
+ "[XXXXX716]",
720
+ "[XXXXX717]",
721
+ "[XXXXX718]",
722
+ "[XXXXX719]",
723
+ "[XXXXX720]",
724
+ "[XXXXX721]",
725
+ "[XXXXX722]",
726
+ "[XXXXX723]",
727
+ "[XXXXX724]",
728
+ "[XXXXX725]",
729
+ "[XXXXX726]",
730
+ "[XXXXX727]",
731
+ "[XXXXX728]",
732
+ "[XXXXX729]",
733
+ "[XXXXX730]",
734
+ "[XXXXX731]",
735
+ "[XXXXX732]",
736
+ "[XXXXX733]",
737
+ "[XXXXX734]",
738
+ "[XXXXX735]",
739
+ "[XXXXX736]",
740
+ "[XXXXX737]",
741
+ "[XXXXX738]",
742
+ "[XXXXX739]",
743
+ "[XXXXX740]",
744
+ "[XXXXX741]",
745
+ "[XXXXX742]",
746
+ "[XXXXX743]",
747
+ "[XXXXX744]",
748
+ "[XXXXX745]",
749
+ "[XXXXX746]",
750
+ "[XXXXX747]",
751
+ "[XXXXX748]",
752
+ "[XXXXX749]",
753
+ "[XXXXX750]",
754
+ "[XXXXX751]",
755
+ "[XXXXX752]",
756
+ "[XXXXX753]",
757
+ "[XXXXX754]",
758
+ "[XXXXX755]",
759
+ "[XXXXX756]",
760
+ "[XXXXX757]",
761
+ "[XXXXX758]",
762
+ "[XXXXX759]",
763
+ "[XXXXX760]",
764
+ "[XXXXX761]",
765
+ "[XXXXX762]",
766
+ "[XXXXX763]",
767
+ "[XXXXX764]",
768
+ "[XXXXX765]",
769
+ "[XXXXX766]",
770
+ "[XXXXX767]",
771
+ "[XXXXX768]",
772
+ "[XXXXX769]",
773
+ "[XXXXX770]",
774
+ "[XXXXX771]",
775
+ "[XXXXX772]",
776
+ "[XXXXX773]",
777
+ "[XXXXX774]",
778
+ "[XXXXX775]",
779
+ "[XXXXX776]",
780
+ "[XXXXX777]",
781
+ "[XXXXX778]",
782
+ "[XXXXX779]",
783
+ "[XXXXX780]",
784
+ "[XXXXX781]",
785
+ "[XXXXX782]",
786
+ "[XXXXX783]",
787
+ "[XXXXX784]",
788
+ "[XXXXX785]",
789
+ "[XXXXX786]",
790
+ "[XXXXX787]",
791
+ "[XXXXX788]",
792
+ "[XXXXX789]",
793
+ "[XXXXX790]",
794
+ "[XXXXX791]",
795
+ "[XXXXX792]",
796
+ "[XXXXX793]",
797
+ "[XXXXX794]",
798
+ "[XXXXX795]",
799
+ "[XXXXX796]",
800
+ "[XXXXX797]",
801
+ "[XXXXX798]",
802
+ "[XXXXX799]",
803
+ "[XXXXX800]",
804
+ "[XXXXX801]",
805
+ "[XXXXX802]",
806
+ "[XXXXX803]",
807
+ "[XXXXX804]",
808
+ "[XXXXX805]",
809
+ "[XXXXX806]",
810
+ "[XXXXX807]",
811
+ "[XXXXX808]",
812
+ "[XXXXX809]",
813
+ "[XXXXX810]",
814
+ "[XXXXX811]",
815
+ "[XXXXX812]",
816
+ "[XXXXX813]",
817
+ "[XXXXX814]",
818
+ "[XXXXX815]",
819
+ "[XXXXX816]",
820
+ "[XXXXX817]",
821
+ "[XXXXX818]",
822
+ "[XXXXX819]",
823
+ "[XXXXX820]",
824
+ "[XXXXX821]",
825
+ "[XXXXX822]",
826
+ "[XXXXX823]",
827
+ "[XXXXX824]",
828
+ "[XXXXX825]",
829
+ "[XXXXX826]",
830
+ "[XXXXX827]",
831
+ "[XXXXX828]",
832
+ "[XXXXX829]",
833
+ "[XXXXX830]",
834
+ "[XXXXX831]",
835
+ "[XXXXX832]",
836
+ "[XXXXX833]",
837
+ "[XXXXX834]",
838
+ "[XXXXX835]",
839
+ "[XXXXX836]",
840
+ "[XXXXX837]",
841
+ "[XXXXX838]",
842
+ "[XXXXX839]",
843
+ "[XXXXX840]",
844
+ "[XXXXX841]",
845
+ "[XXXXX842]",
846
+ "[XXXXX843]",
847
+ "[XXXXX844]",
848
+ "[XXXXX845]",
849
+ "[XXXXX846]",
850
+ "[XXXXX847]",
851
+ "[XXXXX848]",
852
+ "[XXXXX849]",
853
+ "[XXXXX850]",
854
+ "[XXXXX851]",
855
+ "[XXXXX852]",
856
+ "[XXXXX853]",
857
+ "[XXXXX854]",
858
+ "[XXXXX855]",
859
+ "[XXXXX856]",
860
+ "[XXXXX857]",
861
+ "[XXXXX858]",
862
+ "[XXXXX859]",
863
+ "[XXXXX860]",
864
+ "[XXXXX861]",
865
+ "[XXXXX862]",
866
+ "[XXXXX863]",
867
+ "[XXXXX864]",
868
+ "[XXXXX865]",
869
+ "[XXXXX866]",
870
+ "[XXXXX867]",
871
+ "[XXXXX868]",
872
+ "[XXXXX869]",
873
+ "[XXXXX870]",
874
+ "[XXXXX871]",
875
+ "[XXXXX872]",
876
+ "[XXXXX873]",
877
+ "[XXXXX874]",
878
+ "[XXXXX875]",
879
+ "[XXXXX876]",
880
+ "[XXXXX877]",
881
+ "[XXXXX878]",
882
+ "[XXXXX879]",
883
+ "[XXXXX880]",
884
+ "[XXXXX881]",
885
+ "[XXXXX882]",
886
+ "[XXXXX883]",
887
+ "[XXXXX884]",
888
+ "[XXXXX885]",
889
+ "[XXXXX886]",
890
+ "[XXXXX887]",
891
+ "[XXXXX888]",
892
+ "[XXXXX889]",
893
+ "[XXXXX890]",
894
+ "[XXXXX891]",
895
+ "[XXXXX892]",
896
+ "[XXXXX893]",
897
+ "[XXXXX894]",
898
+ "[XXXXX895]",
899
+ "[XXXXX896]",
900
+ "[XXXXX897]",
901
+ "[XXXXX898]",
902
+ "[XXXXX899]",
903
+ "[XXXXX900]",
904
+ "[XXXXX901]",
905
+ "[XXXXX902]",
906
+ "[XXXXX903]",
907
+ "[XXXXX904]",
908
+ "[XXXXX905]",
909
+ "[XXXXX906]",
910
+ "[XXXXX907]",
911
+ "[XXXXX908]",
912
+ "[XXXXX909]",
913
+ "[XXXXX910]",
914
+ "[XXXXX911]",
915
+ "[XXXXX912]",
916
+ "[XXXXX913]",
917
+ "[XXXXX914]",
918
+ "[XXXXX915]",
919
+ "[XXXXX916]",
920
+ "[XXXXX917]",
921
+ "[XXXXX918]",
922
+ "[XXXXX919]",
923
+ "[XXXXX920]",
924
+ "[XXXXX921]",
925
+ "[XXXXX922]",
926
+ "[XXXXX923]",
927
+ "[XXXXX924]",
928
+ "[XXXXX925]",
929
+ "[XXXXX926]",
930
+ "[XXXXX927]",
931
+ "[XXXXX928]",
932
+ "[XXXXX929]",
933
+ "[XXXXX930]",
934
+ "[XXXXX931]",
935
+ "[XXXXX932]",
936
+ "[XXXXX933]",
937
+ "[XXXXX934]",
938
+ "[XXXXX935]",
939
+ "[XXXXX936]",
940
+ "[XXXXX937]",
941
+ "[XXXXX938]",
942
+ "[XXXXX939]",
943
+ "[XXXXX940]",
944
+ "[XXXXX941]",
945
+ "[XXXXX942]",
946
+ "[XXXXX943]",
947
+ "[XXXXX944]",
948
+ "[XXXXX945]",
949
+ "[XXXXX946]",
950
+ "[XXXXX947]",
951
+ "[XXXXX948]",
952
+ "[XXXXX949]",
953
+ "[XXXXX950]",
954
+ "[XXXXX951]",
955
+ "[XXXXX952]",
956
+ "[XXXXX953]",
957
+ "[XXXXX954]",
958
+ "[XXXXX955]",
959
+ "[XXXXX956]",
960
+ "[XXXXX957]",
961
+ "[XXXXX958]",
962
+ "[XXXXX959]",
963
+ "[XXXXX960]",
964
+ "[XXXXX961]",
965
+ "[XXXXX962]",
966
+ "[XXXXX963]",
967
+ "[XXXXX964]",
968
+ "[XXXXX965]",
969
+ "[XXXXX966]",
970
+ "[XXXXX967]",
971
+ "[XXXXX968]",
972
+ "[XXXXX969]",
973
+ "[XXXXX970]",
974
+ "[XXXXX971]",
975
+ "[XXXXX972]",
976
+ "[XXXXX973]",
977
+ "[XXXXX974]",
978
+ "[XXXXX975]",
979
+ "[XXXXX976]",
980
+ "[XXXXX977]",
981
+ "[XXXXX978]",
982
+ "[XXXXX979]",
983
+ "[XXXXX980]",
984
+ "[XXXXX981]",
985
+ "[XXXXX982]",
986
+ "[XXXXX983]",
987
+ "[XXXXX984]",
988
+ "[XXXXX985]",
989
+ "[XXXXX986]",
990
+ "[XXXXX987]",
991
+ "[XXXXX988]",
992
+ "[XXXXX989]",
993
+ "[XXXXX990]",
994
+ "[XXXXX991]",
995
+ "[XXXXX992]",
996
+ "[XXXXX993]",
997
+ "[XXXXX994]",
998
+ "[XXXXX995]",
999
+ "[XXXXX996]",
1000
+ "[XXXXX997]",
1001
+ "[XXXXX998]",
1002
+ "[XXXXX999]",
1003
+ "[XXXXX1000]",
1004
+ "[XXXXX1001]",
1005
+ "[XXXXX1002]",
1006
+ "[XXXXX1003]",
1007
+ "[XXXXX1004]",
1008
+ "[XXXXX1005]",
1009
+ "[XXXXX1006]",
1010
+ "[XXXXX1007]",
1011
+ "[XXXXX1008]",
1012
+ "[XXXXX1009]",
1013
+ "[XXXXX1010]",
1014
+ "[XXXXX1011]",
1015
+ "[XXXXX1012]",
1016
+ "[XXXXX1013]",
1017
+ "[XXXXX1014]",
1018
+ "[XXXXX1015]",
1019
+ "[XXXXX1016]",
1020
+ "[XXXXX1017]",
1021
+ "[XXXXX1018]",
1022
+ "[XXXXX1019]",
1023
+ "[XXXXX1020]",
1024
+ "[XXXXX1021]",
1025
+ "[XXXXX1022]",
1026
+ "[XXXXX1023]",
1027
+ "[XXXXX1024]",
1028
+ "[XXXXX1025]",
1029
+ "[XXXXX1026]",
1030
+ "[XXXXX1027]",
1031
+ "[XXXXX1028]",
1032
+ "[XXXXX1029]",
1033
+ "[XXXXX1030]",
1034
+ "[XXXXX1031]",
1035
+ "[XXXXX1032]",
1036
+ "[XXXXX1033]",
1037
+ "[XXXXX1034]",
1038
+ "[XXXXX1035]",
1039
+ "[XXXXX1036]",
1040
+ "[XXXXX1037]",
1041
+ "[XXXXX1038]",
1042
+ "[XXXXX1039]",
1043
+ "[XXXXX1040]",
1044
+ "[XXXXX1041]",
1045
+ "[XXXXX1042]",
1046
+ "[XXXXX1043]",
1047
+ "[XXXXX1044]",
1048
+ "[XXXXX1045]",
1049
+ "[XXXXX1046]",
1050
+ "[XXXXX1047]",
1051
+ "[XXXXX1048]",
1052
+ "[XXXXX1049]",
1053
+ "[XXXXX1050]",
1054
+ "[XXXXX1051]",
1055
+ "[XXXXX1052]",
1056
+ "[XXXXX1053]",
1057
+ "[XXXXX1054]",
1058
+ "[XXXXX1055]",
1059
+ "[XXXXX1056]",
1060
+ "[XXXXX1057]",
1061
+ "[XXXXX1058]",
1062
+ "[XXXXX1059]",
1063
+ "[XXXXX1060]",
1064
+ "[XXXXX1061]",
1065
+ "[XXXXX1062]",
1066
+ "[XXXXX1063]",
1067
+ "[XXXXX1064]",
1068
+ "[XXXXX1065]",
1069
+ "[XXXXX1066]",
1070
+ "[XXXXX1067]",
1071
+ "[XXXXX1068]",
1072
+ "[XXXXX1069]",
1073
+ "[XXXXX1070]",
1074
+ "[XXXXX1071]",
1075
+ "[XXXXX1072]",
1076
+ "[XXXXX1073]",
1077
+ "[XXXXX1074]",
1078
+ "[XXXXX1075]",
1079
+ "[XXXXX1076]",
1080
+ "[XXXXX1077]",
1081
+ "[XXXXX1078]",
1082
+ "[XXXXX1079]",
1083
+ "[XXXXX1080]",
1084
+ "[XXXXX1081]",
1085
+ "[XXXXX1082]",
1086
+ "[XXXXX1083]",
1087
+ "[XXXXX1084]",
1088
+ "[XXXXX1085]",
1089
+ "[XXXXX1086]",
1090
+ "[XXXXX1087]",
1091
+ "[XXXXX1088]",
1092
+ "[XXXXX1089]",
1093
+ "[XXXXX1090]",
1094
+ "[XXXXX1091]",
1095
+ "[XXXXX1092]",
1096
+ "[XXXXX1093]",
1097
+ "[XXXXX1094]",
1098
+ "[XXXXX1095]",
1099
+ "[XXXXX1096]",
1100
+ "[XXXXX1097]",
1101
+ "[XXXXX1098]",
1102
+ "[XXXXX1099]",
1103
+ "[XXXXX1100]",
1104
+ "[XXXXX1101]",
1105
+ "[XXXXX1102]",
1106
+ "[XXXXX1103]",
1107
+ "[XXXXX1104]",
1108
+ "[XXXXX1105]",
1109
+ "[XXXXX1106]",
1110
+ "[XXXXX1107]",
1111
+ "[XXXXX1108]",
1112
+ "[XXXXX1109]",
1113
+ "[XXXXX1110]",
1114
+ "[XXXXX1111]",
1115
+ "[XXXXX1112]",
1116
+ "[XXXXX1113]",
1117
+ "[XXXXX1114]",
1118
+ "[XXXXX1115]",
1119
+ "[XXXXX1116]",
1120
+ "[XXXXX1117]",
1121
+ "[XXXXX1118]",
1122
+ "[XXXXX1119]",
1123
+ "[XXXXX1120]",
1124
+ "[XXXXX1121]",
1125
+ "[XXXXX1122]",
1126
+ "[XXXXX1123]",
1127
+ "[XXXXX1124]",
1128
+ "[XXXXX1125]",
1129
+ "[XXXXX1126]",
1130
+ "[XXXXX1127]",
1131
+ "[XXXXX1128]",
1132
+ "[XXXXX1129]",
1133
+ "[XXXXX1130]",
1134
+ "[XXXXX1131]",
1135
+ "[XXXXX1132]",
1136
+ "[XXXXX1133]",
1137
+ "[XXXXX1134]",
1138
+ "[XXXXX1135]",
1139
+ "[XXXXX1136]",
1140
+ "[XXXXX1137]",
1141
+ "[XXXXX1138]",
1142
+ "[XXXXX1139]",
1143
+ "[XXXXX1140]",
1144
+ "[XXXXX1141]",
1145
+ "[XXXXX1142]",
1146
+ "[XXXXX1143]",
1147
+ "[XXXXX1144]",
1148
+ "[XXXXX1145]",
1149
+ "[XXXXX1146]",
1150
+ "[XXXXX1147]",
1151
+ "[XXXXX1148]",
1152
+ "[XXXXX1149]",
1153
+ "[XXXXX1150]",
1154
+ "[XXXXX1151]",
1155
+ "[XXXXX1152]",
1156
+ "[XXXXX1153]",
1157
+ "[XXXXX1154]",
1158
+ "[XXXXX1155]",
1159
+ "[XXXXX1156]",
1160
+ "[XXXXX1157]",
1161
+ "[XXXXX1158]",
1162
+ "[XXXXX1159]",
1163
+ "[XXXXX1160]",
1164
+ "[XXXXX1161]",
1165
+ "[XXXXX1162]",
1166
+ "[XXXXX1163]",
1167
+ "[XXXXX1164]",
1168
+ "[XXXXX1165]",
1169
+ "[XXXXX1166]",
1170
+ "[XXXXX1167]",
1171
+ "[XXXXX1168]",
1172
+ "[XXXXX1169]",
1173
+ "[XXXXX1170]",
1174
+ "[XXXXX1171]",
1175
+ "[XXXXX1172]",
1176
+ "[XXXXX1173]",
1177
+ "[XXXXX1174]",
1178
+ "[XXXXX1175]",
1179
+ "[XXXXX1176]",
1180
+ "[XXXXX1177]",
1181
+ "[XXXXX1178]",
1182
+ "[XXXXX1179]",
1183
+ "[XXXXX1180]",
1184
+ "[XXXXX1181]",
1185
+ "[XXXXX1182]",
1186
+ "[XXXXX1183]",
1187
+ "[XXXXX1184]",
1188
+ "[XXXXX1185]",
1189
+ "[XXXXX1186]",
1190
+ "[XXXXX1187]",
1191
+ "[XXXXX1188]",
1192
+ "[XXXXX1189]",
1193
+ "[XXXXX1190]",
1194
+ "[XXXXX1191]",
1195
+ "[XXXXX1192]",
1196
+ "[XXXXX1193]",
1197
+ "[XXXXX1194]",
1198
+ "[XXXXX1195]"
1199
+ ],
1200
+ "bos_token": {
1201
+ "content": "[CLS]",
1202
+ "lstrip": false,
1203
+ "normalized": false,
1204
+ "rstrip": false,
1205
+ "single_word": false
1206
+ },
1207
+ "cls_token": {
1208
+ "content": "[CLS]",
1209
+ "lstrip": false,
1210
+ "normalized": false,
1211
+ "rstrip": false,
1212
+ "single_word": false
1213
+ },
1214
+ "eos_token": {
1215
+ "content": "[SEP]",
1216
+ "lstrip": false,
1217
+ "normalized": false,
1218
+ "rstrip": false,
1219
+ "single_word": false
1220
+ },
1221
+ "mask_token": {
1222
+ "content": "[MASK]",
1223
+ "lstrip": true,
1224
+ "normalized": false,
1225
+ "rstrip": false,
1226
+ "single_word": false
1227
+ },
1228
+ "pad_token": {
1229
+ "content": "<pad>",
1230
+ "lstrip": false,
1231
+ "normalized": false,
1232
+ "rstrip": false,
1233
+ "single_word": false
1234
+ },
1235
+ "sep_token": {
1236
+ "content": "[SEP]",
1237
+ "lstrip": false,
1238
+ "normalized": false,
1239
+ "rstrip": false,
1240
+ "single_word": false
1241
+ },
1242
+ "unk_token": {
1243
+ "content": "<unk>",
1244
+ "lstrip": false,
1245
+ "normalized": false,
1246
+ "rstrip": false,
1247
+ "single_word": false
1248
+ }
1249
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
ud.py ADDED
@@ -0,0 +1,142 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import numpy
2
+ from transformers import TokenClassificationPipeline
3
+
4
+ class BellmanFordTokenClassificationPipeline(TokenClassificationPipeline):
5
+ def __init__(self,**kwargs):
6
+ super().__init__(**kwargs)
7
+ x=self.model.config.label2id
8
+ y=[k for k in x if k.startswith("B-") or not (k.startswith("I-") or k.endswith("|root") or k.find("|l-")>0 or k.find("|r-")>0)]
9
+ self.transition=numpy.full((len(x),len(x)),numpy.nan)
10
+ for k,v in x.items():
11
+ for j in ["I-"+k[2:]] if k.startswith("B-") else [k]+y if k.startswith("I-") else y:
12
+ self.transition[v,x[j]]=0
13
+ def check_model_type(self,supported_models):
14
+ pass
15
+ def postprocess(self,model_outputs,**kwargs):
16
+ if "logits" not in model_outputs:
17
+ return self.postprocess(model_outputs[0],**kwargs)
18
+ m=model_outputs["logits"][0].numpy()
19
+ e=numpy.exp(m-numpy.max(m,axis=-1,keepdims=True))
20
+ z=e/e.sum(axis=-1,keepdims=True)
21
+ for i in range(m.shape[0]-1,0,-1):
22
+ m[i-1]+=numpy.nanmax(m[i]+self.transition,axis=1)
23
+ k=[numpy.nanargmax(m[0]+self.transition[0])]
24
+ for i in range(1,m.shape[0]):
25
+ k.append(numpy.nanargmax(m[i]+self.transition[k[-1]]))
26
+ w=[{"entity":self.model.config.id2label[j],"start":s,"end":e,"score":z[i,j]} for i,((s,e),j) in enumerate(zip(model_outputs["offset_mapping"][0].tolist(),k)) if s<e]
27
+ if "aggregation_strategy" in kwargs and kwargs["aggregation_strategy"]!="none":
28
+ for i,t in reversed(list(enumerate(w))):
29
+ p=t.pop("entity")
30
+ if p.startswith("I-"):
31
+ w[i-1]["score"]=min(w[i-1]["score"],t["score"])
32
+ w[i-1]["end"]=w.pop(i)["end"]
33
+ elif p.startswith("B-"):
34
+ t["entity_group"]=p[2:]
35
+ else:
36
+ t["entity_group"]=p
37
+ for t in w:
38
+ t["text"]=model_outputs["sentence"][t["start"]:t["end"]]
39
+ return w
40
+
41
+ class UniversalDependenciesCausalPipeline(BellmanFordTokenClassificationPipeline):
42
+ def __init__(self,**kwargs):
43
+ kwargs["aggregation_strategy"]="simple"
44
+ super().__init__(**kwargs)
45
+ x=self.model.config.label2id
46
+ self.root=numpy.full((len(x)),numpy.nan)
47
+ self.left_arc=numpy.full((len(x)),numpy.nan)
48
+ self.right_arc=numpy.full((len(x)),numpy.nan)
49
+ for k,v in x.items():
50
+ if k.endswith("|root"):
51
+ self.root[v]=0
52
+ elif k.find("|l-")>0:
53
+ self.left_arc[v]=0
54
+ elif k.find("|r-")>0:
55
+ self.right_arc[v]=0
56
+ def postprocess(self,model_outputs,**kwargs):
57
+ import torch
58
+ if "logits" not in model_outputs:
59
+ return self.postprocess(model_outputs[0],**kwargs)
60
+ m=model_outputs["logits"][0].numpy()
61
+ for i in range(m.shape[0]-1,0,-1):
62
+ m[i-1]+=numpy.nanmax(m[i]+self.transition,axis=1)
63
+ k=[numpy.nanargmax(m[0]+self.transition[0])]
64
+ for i in range(1,m.shape[0]):
65
+ k.append(numpy.nanargmax(m[i]+self.transition[k[-1]]))
66
+ w=[{"entity":self.model.config.id2label[j],"start":s,"end":e} for i,((s,e),j) in enumerate(zip(model_outputs["offset_mapping"][0].tolist(),k)) if s<e]
67
+ for i,t in reversed(list(enumerate(w))):
68
+ p=t.pop("entity")
69
+ if p.startswith("I-"):
70
+ w[i-1]["end"]=max(w.pop(i)["end"],w[i-1]["end"])
71
+ elif i>0 and w[i-1]["end"]>w[i]["start"]:
72
+ w[i-1]["end"]=max(w.pop(i)["end"],w[i-1]["end"])
73
+ elif p.startswith("B-"):
74
+ t["entity_group"]=p[2:]
75
+ else:
76
+ t["entity_group"]=p
77
+ d=[model_outputs["sentence"][t["start"]:t["end"]] for t in w]
78
+ for i in range(len(d)-1,-1,-1):
79
+ if d[i].startswith(" "):
80
+ j=len(d[i])-len(d[i].lstrip())
81
+ d[i]=d[i].lstrip()
82
+ w[i]["start"]+=j
83
+ if d[i].endswith(" "):
84
+ j=len(d[i])-len(d[i].rstrip())
85
+ d[i]=d[i].rstrip()
86
+ w[i]["end"]-=j
87
+ if d[i].strip()=="":
88
+ d.pop(i)
89
+ w.pop(i)
90
+ v=self.tokenizer(d,add_special_tokens=False)
91
+ e=self.model.get_input_embeddings().weight
92
+ m=[]
93
+ for x in v["input_ids"]:
94
+ if x==[]:
95
+ x=[self.tokenizer.unk_token_id]
96
+ m.append(e[x,:].sum(axis=0))
97
+ m.append(e[self.tokenizer.sep_token_id,:])
98
+ m.append(e[self.tokenizer.pad_token_id,:])
99
+ m.append(e[self.tokenizer.cls_token_id,:])
100
+ m=torch.stack(m).to(self.device)
101
+ k=list(range(-1,len(d)+1))
102
+ e=[]
103
+ with torch.no_grad():
104
+ for i in range(len(d)):
105
+ e.append(self.model(inputs_embeds=torch.unsqueeze(m[k+list(range(i,len(d)))+[-2]*i,:],0)).logits[0,-len(d):,:])
106
+ e=torch.stack(e).cpu().numpy()
107
+ for i in range(len(d)):
108
+ for j in range(i):
109
+ e[-j-1,-i-1],e[-i-1,-j-1]=e[-i-1,i-j]+self.left_arc,e[-i-1,i-j]+self.right_arc
110
+ e[-i-1,-i-1]=e[-i-1,0]+self.root
111
+ m,p=numpy.nanmax(e,axis=2),numpy.nanargmax(e,axis=2)
112
+ h=self.chu_liu_edmonds(m)
113
+ z=[i for i,j in enumerate(h) if i==j]
114
+ if len(z)>1:
115
+ k,h=z[numpy.nanargmax(m[z,z])],numpy.nanmin(m)-numpy.nanmax(m)
116
+ m[:,z]+=[[0 if j in z and (i!=j or i==k) else h for i in z] for j in range(m.shape[0])]
117
+ h=self.chu_liu_edmonds(m)
118
+ q=[self.model.config.id2label[p[j,i]].split("|") for i,j in enumerate(h)]
119
+ t=model_outputs["sentence"].replace("\n"," ")
120
+ u="# text = "+t+"\n"
121
+ for i,j in enumerate(d):
122
+ u+="\t".join([str(i+1),j,"_",q[i][0],"_","_" if len(q[i])<3 else "|".join(q[i][1:-1]),str(0 if h[i]==i else h[i]+1),"root" if q[i][-1]=="root" else q[i][-1][2:],"_","_" if i+1<len(d) and w[i]["end"]<w[i+1]["start"] else "SpaceAfter=No"])+"\n"
123
+ return u+"\n"
124
+ def chu_liu_edmonds(self,matrix):
125
+ h=numpy.nanargmax(matrix,axis=0)
126
+ x=[-1 if i==j else j for i,j in enumerate(h)]
127
+ for b in [lambda x,i,j:-1 if i not in x else x[i],lambda x,i,j:-1 if j<0 else x[j]]:
128
+ y=[]
129
+ while x!=y:
130
+ y=list(x)
131
+ for i,j in enumerate(x):
132
+ x[i]=b(x,i,j)
133
+ if max(x)<0:
134
+ return h
135
+ y,x=[i for i,j in enumerate(x) if j==max(x)],[i for i,j in enumerate(x) if j<max(x)]
136
+ z=matrix-numpy.nanmax(matrix,axis=0)
137
+ m=numpy.block([[z[x,:][:,x],numpy.nanmax(z[x,:][:,y],axis=1).reshape(len(x),1)],[numpy.nanmax(z[y,:][:,x],axis=0),numpy.nanmax(z[y,y])]])
138
+ k=[j if i==len(x) else x[j] if j<len(x) else y[numpy.nanargmax(z[y,x[i]])] for i,j in enumerate(self.chu_liu_edmonds(m))]
139
+ h=[j if i in y else k[x.index(i)] for i,j in enumerate(h)]
140
+ i=y[numpy.nanargmax(z[x[k[-1]],y] if k[-1]<len(x) else z[y,y])]
141
+ h[i]=x[k[-1]] if k[-1]<len(x) else i
142
+ return h