Hugging Face's logo

niobures
/
spaCy-misc-langs

Model card Files
xet
Community
niobures commited on
Commit
12b5d9f
·
verified ·
1 Parent(s): 8ef8cd5

spaCy (langs: AF, ENHG, EL, GA, HI, IS, LV, MK, SK, TA, VI, XCL)

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +144 -35
  2. AF/af_udv25_afrikaansafribooms_trf/.gitattributes +32 -0
  3. AF/af_udv25_afrikaansafribooms_trf/LICENSE.txt +3 -0
  4. AF/af_udv25_afrikaansafribooms_trf/README.md +108 -0
  5. AF/af_udv25_afrikaansafribooms_trf/af_udv25_afrikaansafribooms_trf-any-py3-none-any.whl +3 -0
  6. AF/af_udv25_afrikaansafribooms_trf/config.cfg +254 -0
  7. AF/af_udv25_afrikaansafribooms_trf/experimental_char_ner_tokenizer/cfg +13 -0
  8. AF/af_udv25_afrikaansafribooms_trf/experimental_char_ner_tokenizer/model +3 -0
  9. AF/af_udv25_afrikaansafribooms_trf/experimental_char_ner_tokenizer/moves +1 -0
  10. AF/af_udv25_afrikaansafribooms_trf/experimental_edit_tree_lemmatizer/cfg +275 -0
  11. AF/af_udv25_afrikaansafribooms_trf/experimental_edit_tree_lemmatizer/model +3 -0
  12. AF/af_udv25_afrikaansafribooms_trf/experimental_edit_tree_lemmatizer/trees +0 -0
  13. AF/af_udv25_afrikaansafribooms_trf/meta.json +735 -0
  14. AF/af_udv25_afrikaansafribooms_trf/morphologizer/cfg +130 -0
  15. AF/af_udv25_afrikaansafribooms_trf/morphologizer/model +3 -0
  16. AF/af_udv25_afrikaansafribooms_trf/parser/cfg +13 -0
  17. AF/af_udv25_afrikaansafribooms_trf/parser/model +3 -0
  18. AF/af_udv25_afrikaansafribooms_trf/parser/moves +1 -0
  19. AF/af_udv25_afrikaansafribooms_trf/senter/cfg +3 -0
  20. AF/af_udv25_afrikaansafribooms_trf/senter/model +0 -0
  21. AF/af_udv25_afrikaansafribooms_trf/tagger/cfg +101 -0
  22. AF/af_udv25_afrikaansafribooms_trf/tagger/model +3 -0
  23. AF/af_udv25_afrikaansafribooms_trf/transformer/cfg +3 -0
  24. AF/af_udv25_afrikaansafribooms_trf/transformer/model +3 -0
  25. AF/af_udv25_afrikaansafribooms_trf/vocab/key2row +3 -0
  26. AF/af_udv25_afrikaansafribooms_trf/vocab/lookups.bin +3 -0
  27. AF/af_udv25_afrikaansafribooms_trf/vocab/strings.json +3 -0
  28. AF/af_udv25_afrikaansafribooms_trf/vocab/vectors +0 -0
  29. AF/af_udv25_afrikaansafribooms_trf/vocab/vectors.cfg +3 -0
  30. EL/el_cop_sigtyp_trf/.gitattributes +40 -0
  31. EL/el_cop_sigtyp_trf/README.md +0 -0
  32. EL/el_cop_sigtyp_trf/config.cfg +211 -0
  33. EL/el_cop_sigtyp_trf/el_cop_sigtyp_trf-any-py3-none-any.whl +3 -0
  34. EL/el_cop_sigtyp_trf/meta.json +0 -0
  35. EL/el_cop_sigtyp_trf/morphologizer/cfg +0 -0
  36. EL/el_cop_sigtyp_trf/morphologizer/model +3 -0
  37. EL/el_cop_sigtyp_trf/parser/cfg +13 -0
  38. EL/el_cop_sigtyp_trf/parser/model +3 -0
  39. EL/el_cop_sigtyp_trf/parser/moves +1 -0
  40. EL/el_cop_sigtyp_trf/tagger/cfg +0 -0
  41. EL/el_cop_sigtyp_trf/tagger/model +3 -0
  42. EL/el_cop_sigtyp_trf/tokenizer +3 -0
  43. EL/el_cop_sigtyp_trf/trainable_lemmatizer/cfg +479 -0
  44. EL/el_cop_sigtyp_trf/trainable_lemmatizer/model +3 -0
  45. EL/el_cop_sigtyp_trf/trainable_lemmatizer/trees +3 -0
  46. EL/el_cop_sigtyp_trf/transformer/cfg +3 -0
  47. EL/el_cop_sigtyp_trf/transformer/model +3 -0
  48. EL/el_cop_sigtyp_trf/vocab/key2row +3 -0
  49. EL/el_cop_sigtyp_trf/vocab/lookups.bin +3 -0
  50. EL/el_cop_sigtyp_trf/vocab/strings.json +3 -0
.gitattributes CHANGED
@@ -1,35 +1,144 @@
1
- *.7z filter=lfs diff=lfs merge=lfs -text
2
- *.arrow filter=lfs diff=lfs merge=lfs -text
3
- *.bin filter=lfs diff=lfs merge=lfs -text
4
- *.bz2 filter=lfs diff=lfs merge=lfs -text
5
- *.ckpt filter=lfs diff=lfs merge=lfs -text
6
- *.ftz filter=lfs diff=lfs merge=lfs -text
7
- *.gz filter=lfs diff=lfs merge=lfs -text
8
- *.h5 filter=lfs diff=lfs merge=lfs -text
9
- *.joblib filter=lfs diff=lfs merge=lfs -text
10
- *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
- *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
- *.model filter=lfs diff=lfs merge=lfs -text
13
- *.msgpack filter=lfs diff=lfs merge=lfs -text
14
- *.npy filter=lfs diff=lfs merge=lfs -text
15
- *.npz filter=lfs diff=lfs merge=lfs -text
16
- *.onnx filter=lfs diff=lfs merge=lfs -text
17
- *.ot filter=lfs diff=lfs merge=lfs -text
18
- *.parquet filter=lfs diff=lfs merge=lfs -text
19
- *.pb filter=lfs diff=lfs merge=lfs -text
20
- *.pickle filter=lfs diff=lfs merge=lfs -text
21
- *.pkl filter=lfs diff=lfs merge=lfs -text
22
- *.pt filter=lfs diff=lfs merge=lfs -text
23
- *.pth filter=lfs diff=lfs merge=lfs -text
24
- *.rar filter=lfs diff=lfs merge=lfs -text
25
- *.safetensors filter=lfs diff=lfs merge=lfs -text
26
- saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
- *.tar.* filter=lfs diff=lfs merge=lfs -text
28
- *.tar filter=lfs diff=lfs merge=lfs -text
29
- *.tflite filter=lfs diff=lfs merge=lfs -text
30
- *.tgz filter=lfs diff=lfs merge=lfs -text
31
- *.wasm filter=lfs diff=lfs merge=lfs -text
32
- *.xz filter=lfs diff=lfs merge=lfs -text
33
- *.zip filter=lfs diff=lfs merge=lfs -text
34
- *.zst filter=lfs diff=lfs merge=lfs -text
35
- *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bin.* filter=lfs diff=lfs merge=lfs -text
5
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
6
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
7
+ *.ftz filter=lfs diff=lfs merge=lfs -text
8
+ *.gz filter=lfs diff=lfs merge=lfs -text
9
+ *.h5 filter=lfs diff=lfs merge=lfs -text
10
+ *.joblib filter=lfs diff=lfs merge=lfs -text
11
+ *.jsonl filter=lfs diff=lfs merge=lfs -text
12
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
13
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
14
+ *.model filter=lfs diff=lfs merge=lfs -text
15
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
16
+ *.npy filter=lfs diff=lfs merge=lfs -text
17
+ *.npz filter=lfs diff=lfs merge=lfs -text
18
+ *.onnx filter=lfs diff=lfs merge=lfs -text
19
+ *.ot filter=lfs diff=lfs merge=lfs -text
20
+ *.parquet filter=lfs diff=lfs merge=lfs -text
21
+ *.pb filter=lfs diff=lfs merge=lfs -text
22
+ *.pickle filter=lfs diff=lfs merge=lfs -text
23
+ *.pkl filter=lfs diff=lfs merge=lfs -text
24
+ *.pt filter=lfs diff=lfs merge=lfs -text
25
+ *.pth filter=lfs diff=lfs merge=lfs -text
26
+ *.rar filter=lfs diff=lfs merge=lfs -text
27
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
28
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
29
+ *.tar filter=lfs diff=lfs merge=lfs -text
30
+ *.tflite filter=lfs diff=lfs merge=lfs -text
31
+ *.tgz filter=lfs diff=lfs merge=lfs -text
32
+ *.txt filter=lfs diff=lfs merge=lfs -text
33
+ *.wasm filter=lfs diff=lfs merge=lfs -text
34
+ *.whl filter=lfs diff=lfs merge=lfs -text
35
+ *.xz filter=lfs diff=lfs merge=lfs -text
36
+ *.zip filter=lfs diff=lfs merge=lfs -text
37
+ *.zst filter=lfs diff=lfs merge=lfs -text
38
+ *key2row filter=lfs diff=lfs merge=lfs -text
39
+ *strings.json filter=lfs diff=lfs merge=lfs -text
40
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
41
+ model filter=lfs diff=lfs merge=lfs -text
42
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
43
+ vectors filter=lfs diff=lfs merge=lfs -text
44
+ AF/af_udv25_afrikaansafribooms_trf/experimental_char_ner_tokenizer/model filter=lfs diff=lfs merge=lfs -text
45
+ AF/af_udv25_afrikaansafribooms_trf/experimental_edit_tree_lemmatizer/model filter=lfs diff=lfs merge=lfs -text
46
+ AF/af_udv25_afrikaansafribooms_trf/morphologizer/model filter=lfs diff=lfs merge=lfs -text
47
+ AF/af_udv25_afrikaansafribooms_trf/parser/model filter=lfs diff=lfs merge=lfs -text
48
+ AF/af_udv25_afrikaansafribooms_trf/tagger/model filter=lfs diff=lfs merge=lfs -text
49
+ AF/af_udv25_afrikaansafribooms_trf/transformer/model filter=lfs diff=lfs merge=lfs -text
50
+ Early[[:space:]]New[[:space:]]High[[:space:]]German/enhg-ner/ner/model filter=lfs diff=lfs merge=lfs -text
51
+ Early[[:space:]]New[[:space:]]High[[:space:]]German/enhg-ner/transformer/model filter=lfs diff=lfs merge=lfs -text
52
+ EL/el_cop_sigtyp_trf/morphologizer/model filter=lfs diff=lfs merge=lfs -text
53
+ EL/el_cop_sigtyp_trf/parser/model filter=lfs diff=lfs merge=lfs -text
54
+ EL/el_cop_sigtyp_trf/tagger/model filter=lfs diff=lfs merge=lfs -text
55
+ EL/el_cop_sigtyp_trf/trainable_lemmatizer/model filter=lfs diff=lfs merge=lfs -text
56
+ EL/el_cop_sigtyp_trf/trainable_lemmatizer/trees filter=lfs diff=lfs merge=lfs -text
57
+ EL/el_cop_sigtyp_trf/transformer/model filter=lfs diff=lfs merge=lfs -text
58
+ EL/el_core_news_lg/lemmatizer/model filter=lfs diff=lfs merge=lfs -text
59
+ EL/el_core_news_lg/lemmatizer/trees filter=lfs diff=lfs merge=lfs -text
60
+ EL/el_core_news_lg/morphologizer/model filter=lfs diff=lfs merge=lfs -text
61
+ EL/el_core_news_lg/ner/model filter=lfs diff=lfs merge=lfs -text
62
+ EL/el_core_news_lg/parser/model filter=lfs diff=lfs merge=lfs -text
63
+ EL/el_core_news_lg/senter/model filter=lfs diff=lfs merge=lfs -text
64
+ EL/el_core_news_lg/tok2vec/model filter=lfs diff=lfs merge=lfs -text
65
+ EL/el_core_news_lg/vocab/vectors filter=lfs diff=lfs merge=lfs -text
66
+ EL/el_core_news_md/lemmatizer/model filter=lfs diff=lfs merge=lfs -text
67
+ EL/el_core_news_md/lemmatizer/trees filter=lfs diff=lfs merge=lfs -text
68
+ EL/el_core_news_md/morphologizer/model filter=lfs diff=lfs merge=lfs -text
69
+ EL/el_core_news_md/ner/model filter=lfs diff=lfs merge=lfs -text
70
+ EL/el_core_news_md/parser/model filter=lfs diff=lfs merge=lfs -text
71
+ EL/el_core_news_md/senter/model filter=lfs diff=lfs merge=lfs -text
72
+ EL/el_core_news_md/tok2vec/model filter=lfs diff=lfs merge=lfs -text
73
+ EL/el_core_news_md/vocab/vectors filter=lfs diff=lfs merge=lfs -text
74
+ EL/el_core_news_sm-osanseviero/morphologizer/model filter=lfs diff=lfs merge=lfs -text
75
+ EL/el_core_news_sm-osanseviero/ner/model filter=lfs diff=lfs merge=lfs -text
76
+ EL/el_core_news_sm-osanseviero/parser/model filter=lfs diff=lfs merge=lfs -text
77
+ EL/el_core_news_sm-osanseviero/senter/model filter=lfs diff=lfs merge=lfs -text
78
+ EL/el_core_news_sm-osanseviero/tok2vec/model filter=lfs diff=lfs merge=lfs -text
79
+ EL/el_core_news_sm/lemmatizer/model filter=lfs diff=lfs merge=lfs -text
80
+ EL/el_core_news_sm/lemmatizer/trees filter=lfs diff=lfs merge=lfs -text
81
+ EL/el_core_news_sm/morphologizer/model filter=lfs diff=lfs merge=lfs -text
82
+ EL/el_core_news_sm/ner/model filter=lfs diff=lfs merge=lfs -text
83
+ EL/el_core_news_sm/parser/model filter=lfs diff=lfs merge=lfs -text
84
+ EL/el_core_news_sm/senter/model filter=lfs diff=lfs merge=lfs -text
85
+ EL/el_core_news_sm/tok2vec/model filter=lfs diff=lfs merge=lfs -text
86
+ GA/ga_udv25_irishidt_trf/experimental_char_ner_tokenizer/model filter=lfs diff=lfs merge=lfs -text
87
+ GA/ga_udv25_irishidt_trf/experimental_edit_tree_lemmatizer/model filter=lfs diff=lfs merge=lfs -text
88
+ GA/ga_udv25_irishidt_trf/experimental_edit_tree_lemmatizer/trees filter=lfs diff=lfs merge=lfs -text
89
+ GA/ga_udv25_irishidt_trf/morphologizer/model filter=lfs diff=lfs merge=lfs -text
90
+ GA/ga_udv25_irishidt_trf/parser/model filter=lfs diff=lfs merge=lfs -text
91
+ GA/ga_udv25_irishidt_trf/tagger/model filter=lfs diff=lfs merge=lfs -text
92
+ GA/ga_udv25_irishidt_trf/transformer/model filter=lfs diff=lfs merge=lfs -text
93
+ HI/hi_ud_hi_ewt/morphologizer/model filter=lfs diff=lfs merge=lfs -text
94
+ HI/hi_ud_hi_ewt/parser/model filter=lfs diff=lfs merge=lfs -text
95
+ HI/hi_ud_hi_ewt/tagger/model filter=lfs diff=lfs merge=lfs -text
96
+ HI/hi_ud_hi_ewt/tok2vec/model filter=lfs diff=lfs merge=lfs -text
97
+ IS/is_core_web_trf/ner/model filter=lfs diff=lfs merge=lfs -text
98
+ IS/is_core_web_trf/parser/model filter=lfs diff=lfs merge=lfs -text
99
+ IS/is_core_web_trf/tagger/model filter=lfs diff=lfs merge=lfs -text
100
+ IS/is_ner_mim_sm/ner/model filter=lfs diff=lfs merge=lfs -text
101
+ IS/is_ner_mim_sm/tok2vec/model filter=lfs diff=lfs merge=lfs -text
102
+ IS/is_ner_mim_trf/ner/model filter=lfs diff=lfs merge=lfs -text
103
+ IS/is_ud_is_pud/morphologizer/model filter=lfs diff=lfs merge=lfs -text
104
+ IS/is_ud_is_pud/parser/model filter=lfs diff=lfs merge=lfs -text
105
+ IS/is_ud_is_pud/tagger/model filter=lfs diff=lfs merge=lfs -text
106
+ IS/is_ud_is_pud/tok2vec/model filter=lfs diff=lfs merge=lfs -text
107
+ LV/lv_udv25_latvianlvtb_trf/experimental_char_ner_tokenizer/model filter=lfs diff=lfs merge=lfs -text
108
+ LV/lv_udv25_latvianlvtb_trf/experimental_edit_tree_lemmatizer/model filter=lfs diff=lfs merge=lfs -text
109
+ LV/lv_udv25_latvianlvtb_trf/experimental_edit_tree_lemmatizer/trees filter=lfs diff=lfs merge=lfs -text
110
+ LV/lv_udv25_latvianlvtb_trf/morphologizer/model filter=lfs diff=lfs merge=lfs -text
111
+ LV/lv_udv25_latvianlvtb_trf/parser/model filter=lfs diff=lfs merge=lfs -text
112
+ LV/lv_udv25_latvianlvtb_trf/tagger/model filter=lfs diff=lfs merge=lfs -text
113
+ LV/lv_udv25_latvianlvtb_trf/transformer/model filter=lfs diff=lfs merge=lfs -text
114
+ MK/mk_core_news_lg/morphologizer/model filter=lfs diff=lfs merge=lfs -text
115
+ MK/mk_core_news_lg/ner/model filter=lfs diff=lfs merge=lfs -text
116
+ MK/mk_core_news_lg/parser/model filter=lfs diff=lfs merge=lfs -text
117
+ MK/mk_core_news_lg/senter/model filter=lfs diff=lfs merge=lfs -text
118
+ MK/mk_core_news_lg/vocab/vectors filter=lfs diff=lfs merge=lfs -text
119
+ MK/mk_core_news_md/morphologizer/model filter=lfs diff=lfs merge=lfs -text
120
+ MK/mk_core_news_md/ner/model filter=lfs diff=lfs merge=lfs -text
121
+ MK/mk_core_news_md/parser/model filter=lfs diff=lfs merge=lfs -text
122
+ MK/mk_core_news_md/senter/model filter=lfs diff=lfs merge=lfs -text
123
+ MK/mk_core_news_md/vocab/vectors filter=lfs diff=lfs merge=lfs -text
124
+ MK/mk_core_news_sm/morphologizer/model filter=lfs diff=lfs merge=lfs -text
125
+ MK/mk_core_news_sm/ner/model filter=lfs diff=lfs merge=lfs -text
126
+ MK/mk_core_news_sm/parser/model filter=lfs diff=lfs merge=lfs -text
127
+ MK/mk_core_news_sm/senter/model filter=lfs diff=lfs merge=lfs -text
128
+ SK/ner-rent-sk-spacy/model/ner/model filter=lfs diff=lfs merge=lfs -text
129
+ SK/ner-rent-sk-spacy/model/transformer/model filter=lfs diff=lfs merge=lfs -text
130
+ SK/sk_udv25_slovaksnk_trf/experimental_char_ner_tokenizer/model filter=lfs diff=lfs merge=lfs -text
131
+ SK/sk_udv25_slovaksnk_trf/experimental_edit_tree_lemmatizer/model filter=lfs diff=lfs merge=lfs -text
132
+ SK/sk_udv25_slovaksnk_trf/experimental_edit_tree_lemmatizer/trees filter=lfs diff=lfs merge=lfs -text
133
+ SK/sk_udv25_slovaksnk_trf/morphologizer/model filter=lfs diff=lfs merge=lfs -text
134
+ SK/sk_udv25_slovaksnk_trf/parser/model filter=lfs diff=lfs merge=lfs -text
135
+ SK/sk_udv25_slovaksnk_trf/tagger/model filter=lfs diff=lfs merge=lfs -text
136
+ SK/sk_udv25_slovaksnk_trf/transformer/model filter=lfs diff=lfs merge=lfs -text
137
+ TA/ta_Tamil_NER/ner/model filter=lfs diff=lfs merge=lfs -text
138
+ TA/ta_Tamil_NER/transformer/model filter=lfs diff=lfs merge=lfs -text
139
+ VI/vi_udv25_vietnamesevtb_trf/experimental_char_ner_tokenizer/model filter=lfs diff=lfs merge=lfs -text
140
+ VI/vi_udv25_vietnamesevtb_trf/parser/model filter=lfs diff=lfs merge=lfs -text
141
+ VI/vi_udv25_vietnamesevtb_trf/tagger/model filter=lfs diff=lfs merge=lfs -text
142
+ VI/vi_udv25_vietnamesevtb_trf/transformer/model filter=lfs diff=lfs merge=lfs -text
143
+ XCL/xcl_ud_caval/morphologizer/model filter=lfs diff=lfs merge=lfs -text
144
+ XCL/xcl_ud_caval/tok2vec/model filter=lfs diff=lfs merge=lfs -text
AF/af_udv25_afrikaansafribooms_trf/.gitattributes ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bin.* filter=lfs diff=lfs merge=lfs -text
5
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.model filter=lfs diff=lfs merge=lfs -text
12
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
13
+ *.onnx filter=lfs diff=lfs merge=lfs -text
14
+ *.ot filter=lfs diff=lfs merge=lfs -text
15
+ *.parquet filter=lfs diff=lfs merge=lfs -text
16
+ *.pb filter=lfs diff=lfs merge=lfs -text
17
+ *.pt filter=lfs diff=lfs merge=lfs -text
18
+ *.pth filter=lfs diff=lfs merge=lfs -text
19
+ *.rar filter=lfs diff=lfs merge=lfs -text
20
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
21
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
22
+ *.tflite filter=lfs diff=lfs merge=lfs -text
23
+ *.tgz filter=lfs diff=lfs merge=lfs -text
24
+ *.xz filter=lfs diff=lfs merge=lfs -text
25
+ *.zip filter=lfs diff=lfs merge=lfs -text
26
+ *.zstandard filter=lfs diff=lfs merge=lfs -text
27
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
28
+ *.whl filter=lfs diff=lfs merge=lfs -text
29
+ *.npz filter=lfs diff=lfs merge=lfs -text
30
+ *strings.json filter=lfs diff=lfs merge=lfs -text
31
+ vectors filter=lfs diff=lfs merge=lfs -text
32
+ model filter=lfs diff=lfs merge=lfs -text
AF/af_udv25_afrikaansafribooms_trf/LICENSE.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c219395bf819ad073b54261c8774df0ff61e16ae0eacc59474aaeed021fa15ca
3
+ size 206
AF/af_udv25_afrikaansafribooms_trf/README.md ADDED
@@ -0,0 +1,108 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - spacy
4
+ - token-classification
5
+ language:
6
+ - af
7
+ license: cc-by-sa-4.0
8
+ model-index:
9
+ - name: af_udv25_afrikaansafribooms_trf
10
+ results:
11
+ - task:
12
+ name: TAG
13
+ type: token-classification
14
+ metrics:
15
+ - name: TAG (XPOS) Accuracy
16
+ type: accuracy
17
+ value: 0.9601278917
18
+ - task:
19
+ name: POS
20
+ type: token-classification
21
+ metrics:
22
+ - name: POS (UPOS) Accuracy
23
+ type: accuracy
24
+ value: 0.9852374236
25
+ - task:
26
+ name: MORPH
27
+ type: token-classification
28
+ metrics:
29
+ - name: Morph (UFeats) Accuracy
30
+ type: accuracy
31
+ value: 0.9751739703
32
+ - task:
33
+ name: LEMMA
34
+ type: token-classification
35
+ metrics:
36
+ - name: Lemma Accuracy
37
+ type: accuracy
38
+ value: 0.9786593964
39
+ - task:
40
+ name: UNLABELED_DEPENDENCIES
41
+ type: token-classification
42
+ metrics:
43
+ - name: Unlabeled Attachment Score (UAS)
44
+ type: f_score
45
+ value: 0.9078427294
46
+ - task:
47
+ name: LABELED_DEPENDENCIES
48
+ type: token-classification
49
+ metrics:
50
+ - name: Labeled Attachment Score (LAS)
51
+ type: f_score
52
+ value: 0.8749739963
53
+ - task:
54
+ name: SENTS
55
+ type: token-classification
56
+ metrics:
57
+ - name: Sentences F-Score
58
+ type: f_score
59
+ value: 1.0
60
+ ---
61
+ UD v2.5 benchmarking pipeline for UD_Afrikaans-AfriBooms
62
+
63
+ | Feature | Description |
64
+ | --- | --- |
65
+ | **Name** | `af_udv25_afrikaansafribooms_trf` |
66
+ | **Version** | `0.0.1` |
67
+ | **spaCy** | `>=3.2.1,<3.3.0` |
68
+ | **Default Pipeline** | `experimental_char_ner_tokenizer`, `transformer`, `tagger`, `morphologizer`, `parser`, `experimental_edit_tree_lemmatizer` |
69
+ | **Components** | `experimental_char_ner_tokenizer`, `transformer`, `senter`, `tagger`, `morphologizer`, `parser`, `experimental_edit_tree_lemmatizer` |
70
+ | **Vectors** | 0 keys, 0 unique vectors (0 dimensions) |
71
+ | **Sources** | [Universal Dependencies v2.5](https://lindat.mff.cuni.cz/repository/xmlui/handle/11234/1-3105) (Zeman, Daniel; et al.) |
72
+ | **License** | `CC BY-SA 4.0` |
73
+ | **Author** | [Explosion](https://explosion.ai) |
74
+
75
+ ### Label Scheme
76
+
77
+ <details>
78
+
79
+ <summary>View label scheme (455 labels for 6 components)</summary>
80
+
81
+ | Component | Labels |
82
+ | --- | --- |
83
+ | **`experimental_char_ner_tokenizer`** | `TOKEN` |
84
+ | **`senter`** | `I`, `S` |
85
+ | **`tagger`** | `AOA`, `AOP`, `ASA`, `ASP`, `AVA`, `AVP`, `BO`, `BS`, `BV`, `KN`, `KO`, `LB`, `LO`, `NA`, `NEE`, `NM`, `NME`, `NSE`, `NSED`, `NSM`, `PA`, `PB`, `PDHEB`, `PDHEDP`, `PDHENP`, `PDHEW`, `PDMB`, `PDMP`, `PDMW`, `PDOENP`, `PDOEW`, `PDVEB`, `PDVEDP`, `PDVENP`, `PDVEW`, `PEEB`, `PEEDP`, `PEENP`, `PEMB`, `PEMP`, `PEMW`, `PO`, `PTEB`, `PTEDP`, `PTENP`, `PTEW`, `PTMP`, `PV`, `PW`, `RA`, `RK`, `RL`, `RO`, `RS`, `RSF`, `RV`, `RWD`, `SVS`, `THAB`, `THAO`, `THBB`, `THBO`, `THNB`, `THPB`, `THPO`, `TRAB`, `TRAO`, `TRBB`, `UPB`, `UPD`, `UPI`, `UPO`, `UPS`, `UPV`, `UPW`, `UXD`, `VTHOG`, `VTHOK`, `VTHOO`, `VTHOV`, `VTHSG`, `VTHSO`, `VTUOA`, `VTUOM`, `VTUOP`, `VUOT`, `VVHOG`, `VVHOK`, `VVHOO`, `VVUOM`, `VVUOP`, `ZE`, `ZM`, `ZPL`, `ZPR` |
86
+ | **`morphologizer`** | `Definite=Def\|POS=DET\|PronType=Art`, `Number=Sing\|POS=NOUN`, `AdpType=Prep\|POS=ADP`, `AdjType=Attr\|Case=Nom\|Degree=Pos\|POS=ADJ`, `Number=Plur\|POS=NOUN`, `POS=AUX\|Tense=Pres\|VerbForm=Fin,Inf\|VerbType=Cop`, `Definite=Ind\|POS=DET\|PronType=Art`, `POS=NUM`, `POS=PART\|PartType=Inf`, `POS=VERB\|Subcat=Tran\|Tense=Pres\|VerbForm=Fin,Inf`, `POS=PRON\|PronType=Rel`, `POS=AUX\|Tense=Pres\|VerbForm=Fin,Inf\|VerbType=Pas`, `POS=PUNCT`, `POS=CCONJ`, `POS=SCONJ`, `POS=VERB\|Subcat=Intr\|Tense=Pres\|VerbForm=Fin,Inf`, `POS=VERB\|Subcat=Intr\|Tense=Past\|VerbForm=Part`, `POS=AUX\|Tense=Past\|VerbForm=Fin\|VerbType=Pas`, `Degree=Pos\|POS=ADV`, `POS=AUX\|Tense=Pres\|VerbForm=Fin,Inf\|VerbType=Mod`, `POS=DET\|PronType=Ind`, `POS=X`, `Number=Sing\|POS=PROPN`, `POS=PRON\|PronType=Ind`, `POS=PART\|PartType=Neg`, `POS=VERB\|Subcat=Tran\|Tense=Past\|VerbForm=Part`, `AdjType=Pred\|Case=Nom\|Degree=Pos\|POS=ADJ`, `POS=DET\|PronType=Dem`, `Degree=Cmp\|POS=ADV`, `Case=Nom\|Number=Sing\|POS=PRON\|Person=3\|PronType=Prs`, `POS=SYM`, `Case=Acc,Nom\|Number=Plur\|POS=PRON\|Person=1\|PronType=Prs`, `POS=PART\|PartType=Gen`, `Case=Acc\|Number=Sing\|POS=PRON\|Person=2\|PronType=Prs`, `Number=Sing\|POS=PRON\|Person=2\|PronType=Prs\|Reflex=Yes`, `Degree=Sup\|POS=ADV`, `Degree=Dim\|Number=Sing\|POS=NOUN`, `Number=Sing\|POS=PRON\|Person=2\|Poss=Yes\|PronType=Prs`, `POS=PRON\|PronType=Int`, `Number=Plur\|POS=PRON\|Person=1\|Poss=Yes\|PronType=Prs`, `Number=Sing\|POS=PRON\|Person=3\|PronType=Prs\|Reflex=Yes`, `Number=Plur\|POS=PRON\|Person=3\|Poss=Yes\|PronType=Prs`, `AdjType=Attr\|Case=Nom\|Degree=Sup\|POS=ADJ`, `Case=Nom\|Number=Sing\|POS=PRON\|Person=2\|PronType=Prs`, `AdjType=Pred\|Case=Nom\|Degree=Cmp\|POS=ADJ`, `POS=VERB\|Subcat=Prep\|Tense=Pres\|VerbForm=Fin,Inf`, `POS=AUX\|Tense=Pres\|VerbForm=Fin,Inf\|VerbType=Aux`, `Number=Sing\|POS=PRON\|Person=3\|Poss=Yes\|PronType=Prs`, `POS=PRON\|PronType=Rcp`, `POS=AUX\|Tense=Past\|VerbForm=Fin\|VerbType=Mod`, `Case=Acc,Nom\|Number=Plur\|POS=PRON\|Person=3\|PronType=Prs`, `POS=AUX\|Tense=Past\|VerbForm=Fin\|VerbType=Cop`, `Case=Acc\|Number=Sing\|POS=PRON\|Person=3\|PronType=Prs`, `Case=Nom\|Number=Sing\|POS=PRON\|Person=1\|PronType=Prs`, `Number=Sing\|POS=PRON\|Person=1\|Poss=Yes\|PronType=Prs`, `Case=Acc,Nom\|Number=Plur\|POS=PRON\|Person=2\|PronType=Prs`, `Number=Plur\|POS=PRON\|Person=3\|PronType=Prs\|Reflex=Yes`, `AdjType=Attr\|Case=Nom\|Degree=Cmp\|POS=ADJ`, `Number=Plur\|POS=PRON\|Person=1\|PronType=Prs\|Reflex=Yes`, `Case=Acc\|Number=Sing\|POS=PRON\|Person=1\|PronType=Prs`, `AdjType=Pred\|Case=Nom\|Degree=Sup\|POS=ADJ` |
87
+ | **`parser`** | `ROOT`, `advmod`, `amod`, `appos`, `aux`, `aux:pass`, `case`, `cc`, `ccomp`, `compound:prt`, `conj`, `cop`, `dep`, `det`, `flat`, `iobj`, `mark`, `nmod`, `nsubj`, `nsubj:pass`, `nummod`, `obj`, `obl`, `punct`, `xcomp` |
88
+ | **`experimental_edit_tree_lemmatizer`** | `1`, `2`, `4`, `7`, `8`, `10`, `12`, `14`, `16`, `18`, `21`, `24`, `26`, `28`, `31`, `32`, `34`, `37`, `39`, `40`, `42`, `44`, `46`, `47`, `49`, `51`, `53`, `54`, `56`, `57`, `58`, `59`, `61`, `64`, `66`, `68`, `69`, `72`, `74`, `75`, `77`, `78`, `81`, `83`, `84`, `85`, `86`, `87`, `90`, `92`, `94`, `96`, `99`, `101`, `103`, `105`, `108`, `110`, `113`, `116`, `117`, `118`, `121`, `123`, `124`, `125`, `127`, `128`, `129`, `133`, `136`, `138`, `141`, `143`, `145`, `147`, `151`, `153`, `154`, `156`, `158`, `159`, `160`, `162`, `164`, `165`, `167`, `168`, `170`, `172`, `174`, `176`, `178`, `179`, `180`, `181`, `183`, `185`, `189`, `190`, `191`, `192`, `194`, `195`, `197`, `198`, `201`, `202`, `203`, `204`, `206`, `207`, `209`, `213`, `214`, `216`, `217`, `218`, `220`, `221`, `222`, `223`, `225`, `226`, `228`, `229`, `231`, `233`, `234`, `236`, `238`, `240`, `241`, `244`, `247`, `248`, `249`, `250`, `252`, `253`, `255`, `256`, `257`, `258`, `261`, `262`, `263`, `265`, `267`, `269`, `270`, `271`, `273`, `275`, `276`, `278`, `279`, `281`, `283`, `285`, `287`, `289`, `291`, `294`, `296`, `297`, `298`, `299`, `300`, `301`, `302`, `303`, `305`, `306`, `307`, `309`, `310`, `311`, `313`, `314`, `315`, `317`, `320`, `321`, `323`, `325`, `326`, `327`, `328`, `329`, `330`, `332`, `333`, `335`, `336`, `337`, `338`, `339`, `340`, `341`, `343`, `344`, `347`, `348`, `349`, `351`, `353`, `355`, `357`, `359`, `360`, `361`, `362`, `365`, `366`, `367`, `369`, `371`, `373`, `374`, `375`, `377`, `379`, `381`, `383`, `386`, `388`, `390`, `392`, `393`, `395`, `397`, `398`, `400`, `401`, `402`, `403`, `405`, `406`, `408`, `409`, `411`, `412`, `414`, `417`, `215`, `418`, `419`, `420`, `421`, `422`, `424`, `425`, `426`, `427`, `429`, `431`, `432`, `433`, `434`, `436`, `438`, `439`, `440`, `442`, `443`, `444`, `447`, `449`, `450`, `452` |
89
+
90
+ </details>
91
+
92
+ ### Accuracy
93
+
94
+ | Type | Score |
95
+ | --- | --- |
96
+ | `TOKEN_F` | 99.92 |
97
+ | `TOKEN_P` | 99.89 |
98
+ | `TOKEN_R` | 99.94 |
99
+ | `TOKEN_ACC` | 100.00 |
100
+ | `SENTS_F` | 100.00 |
101
+ | `SENTS_P` | 100.00 |
102
+ | `SENTS_R` | 100.00 |
103
+ | `TAG_ACC` | 96.01 |
104
+ | `POS_ACC` | 98.52 |
105
+ | `MORPH_ACC` | 97.52 |
106
+ | `DEP_UAS` | 90.78 |
107
+ | `DEP_LAS` | 87.50 |
108
+ | `LEMMA_ACC` | 97.87 |
AF/af_udv25_afrikaansafribooms_trf/af_udv25_afrikaansafribooms_trf-any-py3-none-any.whl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c162c324d3bd14b68e43091b3b54cd8a4a688592678f4957b3123971eecd2b3
3
+ size 842552429
AF/af_udv25_afrikaansafribooms_trf/config.cfg ADDED
@@ -0,0 +1,254 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [paths]
2
+ train = "corpus/UD_Afrikaans-AfriBooms/train.spacy"
3
+ dev = "corpus/UD_Afrikaans-AfriBooms/dev.spacy"
4
+ vectors = null
5
+ init_tok2vec = null
6
+ tokenizer_source = "training/UD_Afrikaans-AfriBooms/tokenizer/model-best"
7
+ transformer_source = "training/UD_Afrikaans-AfriBooms/transformer/model-best"
8
+
9
+ [system]
10
+ gpu_allocator = "pytorch"
11
+ seed = 0
12
+
13
+ [nlp]
14
+ lang = "af"
15
+ pipeline = ["experimental_char_ner_tokenizer","transformer","senter","tagger","morphologizer","parser","experimental_edit_tree_lemmatizer"]
16
+ batch_size = 64
17
+ disabled = ["senter"]
18
+ before_creation = null
19
+ after_creation = null
20
+ after_pipeline_creation = null
21
+ tokenizer = {"@tokenizers":"spacy-experimental.char_pretokenizer.v1"}
22
+
23
+ [components]
24
+
25
+ [components.experimental_char_ner_tokenizer]
26
+ factory = "experimental_char_ner_tokenizer"
27
+ scorer = {"@scorers":"spacy-experimental.tokenizer_scorer.v1"}
28
+
29
+ [components.experimental_char_ner_tokenizer.model]
30
+ @architectures = "spacy.TransitionBasedParser.v2"
31
+ state_type = "ner"
32
+ extra_state_tokens = false
33
+ hidden_width = 64
34
+ maxout_pieces = 2
35
+ use_upper = true
36
+ nO = null
37
+
38
+ [components.experimental_char_ner_tokenizer.model.tok2vec]
39
+ @architectures = "spacy.Tok2Vec.v2"
40
+
41
+ [components.experimental_char_ner_tokenizer.model.tok2vec.embed]
42
+ @architectures = "spacy.MultiHashEmbed.v2"
43
+ width = 128
44
+ attrs = ["ORTH","LOWER","IS_DIGIT","IS_ALPHA","IS_SPACE","IS_PUNCT"]
45
+ rows = [1000,500,50,50,50,50]
46
+ include_static_vectors = false
47
+
48
+ [components.experimental_char_ner_tokenizer.model.tok2vec.encode]
49
+ @architectures = "spacy.MaxoutWindowEncoder.v2"
50
+ width = 128
51
+ depth = 4
52
+ window_size = 4
53
+ maxout_pieces = 2
54
+
55
+ [components.experimental_edit_tree_lemmatizer]
56
+ factory = "experimental_edit_tree_lemmatizer"
57
+ backoff = "orth"
58
+ min_tree_freq = 1
59
+ overwrite = false
60
+ scorer = {"@scorers":"spacy.lemmatizer_scorer.v1"}
61
+ top_k = 1
62
+
63
+ [components.experimental_edit_tree_lemmatizer.model]
64
+ @architectures = "spacy.Tagger.v1"
65
+ nO = null
66
+
67
+ [components.experimental_edit_tree_lemmatizer.model.tok2vec]
68
+ @architectures = "spacy-transformers.TransformerListener.v1"
69
+ grad_factor = 1.0
70
+ upstream = "transformer"
71
+ pooling = {"@layers":"reduce_mean.v1"}
72
+
73
+ [components.morphologizer]
74
+ factory = "morphologizer"
75
+ extend = false
76
+ overwrite = false
77
+ scorer = {"@scorers":"spacy.morphologizer_scorer.v1"}
78
+
79
+ [components.morphologizer.model]
80
+ @architectures = "spacy.Tagger.v1"
81
+ nO = null
82
+
83
+ [components.morphologizer.model.tok2vec]
84
+ @architectures = "spacy-transformers.TransformerListener.v1"
85
+ grad_factor = 1.0
86
+ upstream = "transformer"
87
+ pooling = {"@layers":"reduce_mean.v1"}
88
+
89
+ [components.parser]
90
+ factory = "parser"
91
+ learn_tokens = false
92
+ min_action_freq = 5
93
+ moves = null
94
+ scorer = {"@scorers":"spacy.parser_scorer.v1"}
95
+ update_with_oracle_cut_size = 100
96
+
97
+ [components.parser.model]
98
+ @architectures = "spacy.TransitionBasedParser.v2"
99
+ state_type = "parser"
100
+ extra_state_tokens = false
101
+ hidden_width = 64
102
+ maxout_pieces = 3
103
+ use_upper = false
104
+ nO = null
105
+
106
+ [components.parser.model.tok2vec]
107
+ @architectures = "spacy-transformers.TransformerListener.v1"
108
+ grad_factor = 1.0
109
+ upstream = "transformer"
110
+ pooling = {"@layers":"reduce_mean.v1"}
111
+
112
+ [components.senter]
113
+ factory = "senter"
114
+ overwrite = false
115
+ scorer = {"@scorers":"spacy.senter_scorer.v1"}
116
+
117
+ [components.senter.model]
118
+ @architectures = "spacy.Tagger.v1"
119
+ nO = null
120
+
121
+ [components.senter.model.tok2vec]
122
+ @architectures = "spacy-transformers.TransformerListener.v1"
123
+ grad_factor = 1.0
124
+ upstream = "transformer"
125
+ pooling = {"@layers":"reduce_mean.v1"}
126
+
127
+ [components.tagger]
128
+ factory = "tagger"
129
+ neg_prefix = "!"
130
+ overwrite = false
131
+ scorer = {"@scorers":"spacy.tagger_scorer.v1"}
132
+
133
+ [components.tagger.model]
134
+ @architectures = "spacy.Tagger.v1"
135
+ nO = null
136
+
137
+ [components.tagger.model.tok2vec]
138
+ @architectures = "spacy-transformers.TransformerListener.v1"
139
+ grad_factor = 1.0
140
+ upstream = "transformer"
141
+ pooling = {"@layers":"reduce_mean.v1"}
142
+
143
+ [components.transformer]
144
+ factory = "transformer"
145
+ max_batch_items = 4096
146
+ set_extra_annotations = {"@annotation_setters":"spacy-transformers.null_annotation_setter.v1"}
147
+
148
+ [components.transformer.model]
149
+ @architectures = "spacy-transformers.TransformerModel.v3"
150
+ name = "xlm-roberta-base"
151
+ mixed_precision = true
152
+
153
+ [components.transformer.model.get_spans]
154
+ @span_getters = "spacy-transformers.strided_spans.v1"
155
+ window = 128
156
+ stride = 96
157
+
158
+ [components.transformer.model.grad_scaler_config]
159
+
160
+ [components.transformer.model.tokenizer_config]
161
+ use_fast = true
162
+
163
+ [components.transformer.model.transformer_config]
164
+
165
+ [corpora]
166
+
167
+ [corpora.dev]
168
+ @readers = "spacy.Corpus.v1"
169
+ path = ${paths.dev}
170
+ max_length = 0
171
+ gold_preproc = false
172
+ limit = 0
173
+ augmenter = null
174
+
175
+ [corpora.train]
176
+ @readers = "spacy.Corpus.v1"
177
+ path = ${paths.train}
178
+ max_length = 0
179
+ gold_preproc = false
180
+ limit = 0
181
+ augmenter = null
182
+
183
+ [training]
184
+ train_corpus = "corpora.train"
185
+ dev_corpus = "corpora.dev"
186
+ seed = ${system:seed}
187
+ gpu_allocator = ${system:gpu_allocator}
188
+ dropout = 0.1
189
+ accumulate_gradient = 3
190
+ patience = 5000
191
+ max_epochs = 0
192
+ max_steps = 20000
193
+ eval_frequency = 200
194
+ frozen_components = []
195
+ before_to_disk = null
196
+ annotating_components = []
197
+
198
+ [training.batcher]
199
+ @batchers = "spacy.batch_by_padded.v1"
200
+ discard_oversize = true
201
+ get_length = null
202
+ size = 2000
203
+ buffer = 256
204
+
205
+ [training.logger]
206
+ @loggers = "spacy.ConsoleLogger.v1"
207
+ progress_bar = false
208
+
209
+ [training.optimizer]
210
+ @optimizers = "Adam.v1"
211
+ beta1 = 0.9
212
+ beta2 = 0.999
213
+ L2_is_weight_decay = true
214
+ L2 = 0.01
215
+ grad_clip = 1.0
216
+ use_averages = true
217
+ eps = 0.00000001
218
+
219
+ [training.optimizer.learn_rate]
220
+ @schedules = "warmup_linear.v1"
221
+ warmup_steps = 250
222
+ total_steps = 20000
223
+ initial_rate = 0.00005
224
+
225
+ [training.score_weights]
226
+ token_f = 0.0
227
+ token_p = null
228
+ token_r = null
229
+ token_acc = null
230
+ sents_f = 0.05
231
+ sents_p = 0.0
232
+ sents_r = 0.0
233
+ tag_acc = 0.11
234
+ pos_acc = 0.05
235
+ morph_acc = 0.05
236
+ morph_per_feat = null
237
+ dep_uas = 0.11
238
+ dep_las = 0.11
239
+ dep_las_per_type = null
240
+ lemma_acc = 0.52
241
+
242
+ [pretraining]
243
+
244
+ [initialize]
245
+ vectors = ${paths.vectors}
246
+ init_tok2vec = ${paths.init_tok2vec}
247
+ vocab_data = null
248
+ lookups = null
249
+ before_init = null
250
+ after_init = null
251
+
252
+ [initialize.components]
253
+
254
+ [initialize.tokenizer]
AF/af_udv25_afrikaansafribooms_trf/experimental_char_ner_tokenizer/cfg ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "moves":null,
3
+ "update_with_oracle_cut_size":100,
4
+ "multitasks":[
5
+
6
+ ],
7
+ "min_action_freq":1,
8
+ "learn_tokens":false,
9
+ "beam_width":1,
10
+ "beam_density":0.0,
11
+ "beam_update_prob":0.0,
12
+ "incorrect_spans_key":null
13
+ }
AF/af_udv25_afrikaansafribooms_trf/experimental_char_ner_tokenizer/model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d1451ad867d113c07a1655e5bc3c2cf2b32a5413cac9998fb65e48f36cf083d
3
+ size 6922248
AF/af_udv25_afrikaansafribooms_trf/experimental_char_ner_tokenizer/moves ADDED
@@ -0,0 +1 @@
 
 
1
+ ��moves�l{"0":{},"1":{"TOKEN":168807},"2":{"TOKEN":168807},"3":{"TOKEN":168807},"4":{"TOKEN":168807,"":1},"5":{"":1}}�cfg��neg_key�
AF/af_udv25_afrikaansafribooms_trf/experimental_edit_tree_lemmatizer/cfg ADDED
@@ -0,0 +1,275 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "labels":[
3
+ 1,
4
+ 2,
5
+ 4,
6
+ 7,
7
+ 8,
8
+ 10,
9
+ 12,
10
+ 14,
11
+ 16,
12
+ 18,
13
+ 21,
14
+ 24,
15
+ 26,
16
+ 28,
17
+ 31,
18
+ 32,
19
+ 34,
20
+ 37,
21
+ 39,
22
+ 40,
23
+ 42,
24
+ 44,
25
+ 46,
26
+ 47,
27
+ 49,
28
+ 51,
29
+ 53,
30
+ 54,
31
+ 56,
32
+ 57,
33
+ 58,
34
+ 59,
35
+ 61,
36
+ 64,
37
+ 66,
38
+ 68,
39
+ 69,
40
+ 72,
41
+ 74,
42
+ 75,
43
+ 77,
44
+ 78,
45
+ 81,
46
+ 83,
47
+ 84,
48
+ 85,
49
+ 86,
50
+ 87,
51
+ 90,
52
+ 92,
53
+ 94,
54
+ 96,
55
+ 99,
56
+ 101,
57
+ 103,
58
+ 105,
59
+ 108,
60
+ 110,
61
+ 113,
62
+ 116,
63
+ 117,
64
+ 118,
65
+ 121,
66
+ 123,
67
+ 124,
68
+ 125,
69
+ 127,
70
+ 128,
71
+ 129,
72
+ 133,
73
+ 136,
74
+ 138,
75
+ 141,
76
+ 143,
77
+ 145,
78
+ 147,
79
+ 151,
80
+ 153,
81
+ 154,
82
+ 156,
83
+ 158,
84
+ 159,
85
+ 160,
86
+ 162,
87
+ 164,
88
+ 165,
89
+ 167,
90
+ 168,
91
+ 170,
92
+ 172,
93
+ 174,
94
+ 176,
95
+ 178,
96
+ 179,
97
+ 180,
98
+ 181,
99
+ 183,
100
+ 185,
101
+ 189,
102
+ 190,
103
+ 191,
104
+ 192,
105
+ 194,
106
+ 195,
107
+ 197,
108
+ 198,
109
+ 201,
110
+ 202,
111
+ 203,
112
+ 204,
113
+ 206,
114
+ 207,
115
+ 209,
116
+ 213,
117
+ 214,
118
+ 216,
119
+ 217,
120
+ 218,
121
+ 220,
122
+ 221,
123
+ 222,
124
+ 223,
125
+ 225,
126
+ 226,
127
+ 228,
128
+ 229,
129
+ 231,
130
+ 233,
131
+ 234,
132
+ 236,
133
+ 238,
134
+ 240,
135
+ 241,
136
+ 244,
137
+ 247,
138
+ 248,
139
+ 249,
140
+ 250,
141
+ 252,
142
+ 253,
143
+ 255,
144
+ 256,
145
+ 257,
146
+ 258,
147
+ 261,
148
+ 262,
149
+ 263,
150
+ 265,
151
+ 267,
152
+ 269,
153
+ 270,
154
+ 271,
155
+ 273,
156
+ 275,
157
+ 276,
158
+ 278,
159
+ 279,
160
+ 281,
161
+ 283,
162
+ 285,
163
+ 287,
164
+ 289,
165
+ 291,
166
+ 294,
167
+ 296,
168
+ 297,
169
+ 298,
170
+ 299,
171
+ 300,
172
+ 301,
173
+ 302,
174
+ 303,
175
+ 305,
176
+ 306,
177
+ 307,
178
+ 309,
179
+ 310,
180
+ 311,
181
+ 313,
182
+ 314,
183
+ 315,
184
+ 317,
185
+ 320,
186
+ 321,
187
+ 323,
188
+ 325,
189
+ 326,
190
+ 327,
191
+ 328,
192
+ 329,
193
+ 330,
194
+ 332,
195
+ 333,
196
+ 335,
197
+ 336,
198
+ 337,
199
+ 338,
200
+ 339,
201
+ 340,
202
+ 341,
203
+ 343,
204
+ 344,
205
+ 347,
206
+ 348,
207
+ 349,
208
+ 351,
209
+ 353,
210
+ 355,
211
+ 357,
212
+ 359,
213
+ 360,
214
+ 361,
215
+ 362,
216
+ 365,
217
+ 366,
218
+ 367,
219
+ 369,
220
+ 371,
221
+ 373,
222
+ 374,
223
+ 375,
224
+ 377,
225
+ 379,
226
+ 381,
227
+ 383,
228
+ 386,
229
+ 388,
230
+ 390,
231
+ 392,
232
+ 393,
233
+ 395,
234
+ 397,
235
+ 398,
236
+ 400,
237
+ 401,
238
+ 402,
239
+ 403,
240
+ 405,
241
+ 406,
242
+ 408,
243
+ 409,
244
+ 411,
245
+ 412,
246
+ 414,
247
+ 417,
248
+ 215,
249
+ 418,
250
+ 419,
251
+ 420,
252
+ 421,
253
+ 422,
254
+ 424,
255
+ 425,
256
+ 426,
257
+ 427,
258
+ 429,
259
+ 431,
260
+ 432,
261
+ 433,
262
+ 434,
263
+ 436,
264
+ 438,
265
+ 439,
266
+ 440,
267
+ 442,
268
+ 443,
269
+ 444,
270
+ 447,
271
+ 449,
272
+ 450,
273
+ 452
274
+ ]
275
+ }
AF/af_udv25_afrikaansafribooms_trf/experimental_edit_tree_lemmatizer/model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f440416add1f6b780a89284f4bdfee9ba38635a91fd347722b1963e21652d90
3
+ size 834197
AF/af_udv25_afrikaansafribooms_trf/experimental_edit_tree_lemmatizer/trees ADDED
Binary file (36.1 kB). View file
 
AF/af_udv25_afrikaansafribooms_trf/meta.json ADDED
@@ -0,0 +1,735 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "lang":"af",
3
+ "name":"udv25_afrikaansafribooms_trf",
4
+ "version":"0.0.1",
5
+ "description":"UD v2.5 benchmarking pipeline for UD_Afrikaans-AfriBooms",
6
+ "author":"Explosion",
7
+ "email":"contact@explosion.ai",
8
+ "url":"https://explosion.ai",
9
+ "license":"CC BY-SA 4.0",
10
+ "spacy_version":">=3.2.1,<3.3.0",
11
+ "spacy_git_version":"800737b41",
12
+ "vectors":{
13
+ "width":0,
14
+ "vectors":0,
15
+ "keys":0,
16
+ "name":null
17
+ },
18
+ "labels":{
19
+ "experimental_char_ner_tokenizer":[
20
+ "TOKEN"
21
+ ],
22
+ "transformer":[
23
+
24
+ ],
25
+ "senter":[
26
+ "I",
27
+ "S"
28
+ ],
29
+ "tagger":[
30
+ "AOA",
31
+ "AOP",
32
+ "ASA",
33
+ "ASP",
34
+ "AVA",
35
+ "AVP",
36
+ "BO",
37
+ "BS",
38
+ "BV",
39
+ "KN",
40
+ "KO",
41
+ "LB",
42
+ "LO",
43
+ "NA",
44
+ "NEE",
45
+ "NM",
46
+ "NME",
47
+ "NSE",
48
+ "NSED",
49
+ "NSM",
50
+ "PA",
51
+ "PB",
52
+ "PDHEB",
53
+ "PDHEDP",
54
+ "PDHENP",
55
+ "PDHEW",
56
+ "PDMB",
57
+ "PDMP",
58
+ "PDMW",
59
+ "PDOENP",
60
+ "PDOEW",
61
+ "PDVEB",
62
+ "PDVEDP",
63
+ "PDVENP",
64
+ "PDVEW",
65
+ "PEEB",
66
+ "PEEDP",
67
+ "PEENP",
68
+ "PEMB",
69
+ "PEMP",
70
+ "PEMW",
71
+ "PO",
72
+ "PTEB",
73
+ "PTEDP",
74
+ "PTENP",
75
+ "PTEW",
76
+ "PTMP",
77
+ "PV",
78
+ "PW",
79
+ "RA",
80
+ "RK",
81
+ "RL",
82
+ "RO",
83
+ "RS",
84
+ "RSF",
85
+ "RV",
86
+ "RWD",
87
+ "SVS",
88
+ "THAB",
89
+ "THAO",
90
+ "THBB",
91
+ "THBO",
92
+ "THNB",
93
+ "THPB",
94
+ "THPO",
95
+ "TRAB",
96
+ "TRAO",
97
+ "TRBB",
98
+ "UPB",
99
+ "UPD",
100
+ "UPI",
101
+ "UPO",
102
+ "UPS",
103
+ "UPV",
104
+ "UPW",
105
+ "UXD",
106
+ "VTHOG",
107
+ "VTHOK",
108
+ "VTHOO",
109
+ "VTHOV",
110
+ "VTHSG",
111
+ "VTHSO",
112
+ "VTUOA",
113
+ "VTUOM",
114
+ "VTUOP",
115
+ "VUOT",
116
+ "VVHOG",
117
+ "VVHOK",
118
+ "VVHOO",
119
+ "VVUOM",
120
+ "VVUOP",
121
+ "ZE",
122
+ "ZM",
123
+ "ZPL",
124
+ "ZPR"
125
+ ],
126
+ "morphologizer":[
127
+ "Definite=Def|POS=DET|PronType=Art",
128
+ "Number=Sing|POS=NOUN",
129
+ "AdpType=Prep|POS=ADP",
130
+ "AdjType=Attr|Case=Nom|Degree=Pos|POS=ADJ",
131
+ "Number=Plur|POS=NOUN",
132
+ "POS=AUX|Tense=Pres|VerbForm=Fin,Inf|VerbType=Cop",
133
+ "Definite=Ind|POS=DET|PronType=Art",
134
+ "POS=NUM",
135
+ "POS=PART|PartType=Inf",
136
+ "POS=VERB|Subcat=Tran|Tense=Pres|VerbForm=Fin,Inf",
137
+ "POS=PRON|PronType=Rel",
138
+ "POS=AUX|Tense=Pres|VerbForm=Fin,Inf|VerbType=Pas",
139
+ "POS=PUNCT",
140
+ "POS=CCONJ",
141
+ "POS=SCONJ",
142
+ "POS=VERB|Subcat=Intr|Tense=Pres|VerbForm=Fin,Inf",
143
+ "POS=VERB|Subcat=Intr|Tense=Past|VerbForm=Part",
144
+ "POS=AUX|Tense=Past|VerbForm=Fin|VerbType=Pas",
145
+ "Degree=Pos|POS=ADV",
146
+ "POS=AUX|Tense=Pres|VerbForm=Fin,Inf|VerbType=Mod",
147
+ "POS=DET|PronType=Ind",
148
+ "POS=X",
149
+ "Number=Sing|POS=PROPN",
150
+ "POS=PRON|PronType=Ind",
151
+ "POS=PART|PartType=Neg",
152
+ "POS=VERB|Subcat=Tran|Tense=Past|VerbForm=Part",
153
+ "AdjType=Pred|Case=Nom|Degree=Pos|POS=ADJ",
154
+ "POS=DET|PronType=Dem",
155
+ "Degree=Cmp|POS=ADV",
156
+ "Case=Nom|Number=Sing|POS=PRON|Person=3|PronType=Prs",
157
+ "POS=SYM",
158
+ "Case=Acc,Nom|Number=Plur|POS=PRON|Person=1|PronType=Prs",
159
+ "POS=PART|PartType=Gen",
160
+ "Case=Acc|Number=Sing|POS=PRON|Person=2|PronType=Prs",
161
+ "Number=Sing|POS=PRON|Person=2|PronType=Prs|Reflex=Yes",
162
+ "Degree=Sup|POS=ADV",
163
+ "Degree=Dim|Number=Sing|POS=NOUN",
164
+ "Number=Sing|POS=PRON|Person=2|Poss=Yes|PronType=Prs",
165
+ "POS=PRON|PronType=Int",
166
+ "Number=Plur|POS=PRON|Person=1|Poss=Yes|PronType=Prs",
167
+ "Number=Sing|POS=PRON|Person=3|PronType=Prs|Reflex=Yes",
168
+ "Number=Plur|POS=PRON|Person=3|Poss=Yes|PronType=Prs",
169
+ "AdjType=Attr|Case=Nom|Degree=Sup|POS=ADJ",
170
+ "Case=Nom|Number=Sing|POS=PRON|Person=2|PronType=Prs",
171
+ "AdjType=Pred|Case=Nom|Degree=Cmp|POS=ADJ",
172
+ "POS=VERB|Subcat=Prep|Tense=Pres|VerbForm=Fin,Inf",
173
+ "POS=AUX|Tense=Pres|VerbForm=Fin,Inf|VerbType=Aux",
174
+ "Number=Sing|POS=PRON|Person=3|Poss=Yes|PronType=Prs",
175
+ "POS=PRON|PronType=Rcp",
176
+ "POS=AUX|Tense=Past|VerbForm=Fin|VerbType=Mod",
177
+ "Case=Acc,Nom|Number=Plur|POS=PRON|Person=3|PronType=Prs",
178
+ "POS=AUX|Tense=Past|VerbForm=Fin|VerbType=Cop",
179
+ "Case=Acc|Number=Sing|POS=PRON|Person=3|PronType=Prs",
180
+ "Case=Nom|Number=Sing|POS=PRON|Person=1|PronType=Prs",
181
+ "Number=Sing|POS=PRON|Person=1|Poss=Yes|PronType=Prs",
182
+ "Case=Acc,Nom|Number=Plur|POS=PRON|Person=2|PronType=Prs",
183
+ "Number=Plur|POS=PRON|Person=3|PronType=Prs|Reflex=Yes",
184
+ "AdjType=Attr|Case=Nom|Degree=Cmp|POS=ADJ",
185
+ "Number=Plur|POS=PRON|Person=1|PronType=Prs|Reflex=Yes",
186
+ "Case=Acc|Number=Sing|POS=PRON|Person=1|PronType=Prs",
187
+ "AdjType=Pred|Case=Nom|Degree=Sup|POS=ADJ"
188
+ ],
189
+ "parser":[
190
+ "ROOT",
191
+ "advmod",
192
+ "amod",
193
+ "appos",
194
+ "aux",
195
+ "aux:pass",
196
+ "case",
197
+ "cc",
198
+ "ccomp",
199
+ "compound:prt",
200
+ "conj",
201
+ "cop",
202
+ "dep",
203
+ "det",
204
+ "flat",
205
+ "iobj",
206
+ "mark",
207
+ "nmod",
208
+ "nsubj",
209
+ "nsubj:pass",
210
+ "nummod",
211
+ "obj",
212
+ "obl",
213
+ "punct",
214
+ "xcomp"
215
+ ],
216
+ "experimental_edit_tree_lemmatizer":[
217
+ 1,
218
+ 2,
219
+ 4,
220
+ 7,
221
+ 8,
222
+ 10,
223
+ 12,
224
+ 14,
225
+ 16,
226
+ 18,
227
+ 21,
228
+ 24,
229
+ 26,
230
+ 28,
231
+ 31,
232
+ 32,
233
+ 34,
234
+ 37,
235
+ 39,
236
+ 40,
237
+ 42,
238
+ 44,
239
+ 46,
240
+ 47,
241
+ 49,
242
+ 51,
243
+ 53,
244
+ 54,
245
+ 56,
246
+ 57,
247
+ 58,
248
+ 59,
249
+ 61,
250
+ 64,
251
+ 66,
252
+ 68,
253
+ 69,
254
+ 72,
255
+ 74,
256
+ 75,
257
+ 77,
258
+ 78,
259
+ 81,
260
+ 83,
261
+ 84,
262
+ 85,
263
+ 86,
264
+ 87,
265
+ 90,
266
+ 92,
267
+ 94,
268
+ 96,
269
+ 99,
270
+ 101,
271
+ 103,
272
+ 105,
273
+ 108,
274
+ 110,
275
+ 113,
276
+ 116,
277
+ 117,
278
+ 118,
279
+ 121,
280
+ 123,
281
+ 124,
282
+ 125,
283
+ 127,
284
+ 128,
285
+ 129,
286
+ 133,
287
+ 136,
288
+ 138,
289
+ 141,
290
+ 143,
291
+ 145,
292
+ 147,
293
+ 151,
294
+ 153,
295
+ 154,
296
+ 156,
297
+ 158,
298
+ 159,
299
+ 160,
300
+ 162,
301
+ 164,
302
+ 165,
303
+ 167,
304
+ 168,
305
+ 170,
306
+ 172,
307
+ 174,
308
+ 176,
309
+ 178,
310
+ 179,
311
+ 180,
312
+ 181,
313
+ 183,
314
+ 185,
315
+ 189,
316
+ 190,
317
+ 191,
318
+ 192,
319
+ 194,
320
+ 195,
321
+ 197,
322
+ 198,
323
+ 201,
324
+ 202,
325
+ 203,
326
+ 204,
327
+ 206,
328
+ 207,
329
+ 209,
330
+ 213,
331
+ 214,
332
+ 216,
333
+ 217,
334
+ 218,
335
+ 220,
336
+ 221,
337
+ 222,
338
+ 223,
339
+ 225,
340
+ 226,
341
+ 228,
342
+ 229,
343
+ 231,
344
+ 233,
345
+ 234,
346
+ 236,
347
+ 238,
348
+ 240,
349
+ 241,
350
+ 244,
351
+ 247,
352
+ 248,
353
+ 249,
354
+ 250,
355
+ 252,
356
+ 253,
357
+ 255,
358
+ 256,
359
+ 257,
360
+ 258,
361
+ 261,
362
+ 262,
363
+ 263,
364
+ 265,
365
+ 267,
366
+ 269,
367
+ 270,
368
+ 271,
369
+ 273,
370
+ 275,
371
+ 276,
372
+ 278,
373
+ 279,
374
+ 281,
375
+ 283,
376
+ 285,
377
+ 287,
378
+ 289,
379
+ 291,
380
+ 294,
381
+ 296,
382
+ 297,
383
+ 298,
384
+ 299,
385
+ 300,
386
+ 301,
387
+ 302,
388
+ 303,
389
+ 305,
390
+ 306,
391
+ 307,
392
+ 309,
393
+ 310,
394
+ 311,
395
+ 313,
396
+ 314,
397
+ 315,
398
+ 317,
399
+ 320,
400
+ 321,
401
+ 323,
402
+ 325,
403
+ 326,
404
+ 327,
405
+ 328,
406
+ 329,
407
+ 330,
408
+ 332,
409
+ 333,
410
+ 335,
411
+ 336,
412
+ 337,
413
+ 338,
414
+ 339,
415
+ 340,
416
+ 341,
417
+ 343,
418
+ 344,
419
+ 347,
420
+ 348,
421
+ 349,
422
+ 351,
423
+ 353,
424
+ 355,
425
+ 357,
426
+ 359,
427
+ 360,
428
+ 361,
429
+ 362,
430
+ 365,
431
+ 366,
432
+ 367,
433
+ 369,
434
+ 371,
435
+ 373,
436
+ 374,
437
+ 375,
438
+ 377,
439
+ 379,
440
+ 381,
441
+ 383,
442
+ 386,
443
+ 388,
444
+ 390,
445
+ 392,
446
+ 393,
447
+ 395,
448
+ 397,
449
+ 398,
450
+ 400,
451
+ 401,
452
+ 402,
453
+ 403,
454
+ 405,
455
+ 406,
456
+ 408,
457
+ 409,
458
+ 411,
459
+ 412,
460
+ 414,
461
+ 417,
462
+ 215,
463
+ 418,
464
+ 419,
465
+ 420,
466
+ 421,
467
+ 422,
468
+ 424,
469
+ 425,
470
+ 426,
471
+ 427,
472
+ 429,
473
+ 431,
474
+ 432,
475
+ 433,
476
+ 434,
477
+ 436,
478
+ 438,
479
+ 439,
480
+ 440,
481
+ 442,
482
+ 443,
483
+ 444,
484
+ 447,
485
+ 449,
486
+ 450,
487
+ 452
488
+ ]
489
+ },
490
+ "pipeline":[
491
+ "experimental_char_ner_tokenizer",
492
+ "transformer",
493
+ "tagger",
494
+ "morphologizer",
495
+ "parser",
496
+ "experimental_edit_tree_lemmatizer"
497
+ ],
498
+ "components":[
499
+ "experimental_char_ner_tokenizer",
500
+ "transformer",
501
+ "senter",
502
+ "tagger",
503
+ "morphologizer",
504
+ "parser",
505
+ "experimental_edit_tree_lemmatizer"
506
+ ],
507
+ "disabled":[
508
+ "senter"
509
+ ],
510
+ "sources":[
511
+ {
512
+ "name":"Universal Dependencies v2.5",
513
+ "url":"https://lindat.mff.cuni.cz/repository/xmlui/handle/11234/1-3105",
514
+ "author":"Zeman, Daniel; et al."
515
+ }
516
+ ],
517
+ "performance":{
518
+ "token_f":0.9991538968,
519
+ "token_p":0.9988721805,
520
+ "token_r":0.9994357721,
521
+ "token_acc":1.0,
522
+ "sents_f":1.0,
523
+ "sents_p":1.0,
524
+ "sents_r":1.0,
525
+ "tag_acc":0.9601278917,
526
+ "pos_acc":0.9852374236,
527
+ "morph_acc":0.9751739703,
528
+ "morph_per_feat":{
529
+ "Case":{
530
+ "p":0.9530332681,
531
+ "r":0.9779116466,
532
+ "f":0.9653121903
533
+ },
534
+ "Number":{
535
+ "p":0.9856733524,
536
+ "r":0.9842632332,
537
+ "f":0.9849677881
538
+ },
539
+ "Person":{
540
+ "p":0.9866666667,
541
+ "r":0.9910714286,
542
+ "f":0.9888641425
543
+ },
544
+ "PronType":{
545
+ "p":0.993877551,
546
+ "r":0.9868287741,
547
+ "f":0.9903406202
548
+ },
549
+ "Subcat":{
550
+ "p":0.9416846652,
551
+ "r":0.9256900212,
552
+ "f":0.9336188437
553
+ },
554
+ "Tense":{
555
+ "p":0.992481203,
556
+ "r":0.982630273,
557
+ "f":0.9875311721
558
+ },
559
+ "VerbForm":{
560
+ "p":0.9912280702,
561
+ "r":0.9813895782,
562
+ "f":0.9862842893
563
+ },
564
+ "AdjType":{
565
+ "p":0.9323943662,
566
+ "r":0.9764011799,
567
+ "f":0.9538904899
568
+ },
569
+ "Degree":{
570
+ "p":0.9627659574,
571
+ "r":0.9679144385,
572
+ "f":0.9653333333
573
+ },
574
+ "AdpType":{
575
+ "p":0.9967845659,
576
+ "r":0.9967845659,
577
+ "f":0.9967845659
578
+ },
579
+ "Definite":{
580
+ "p":1.0,
581
+ "r":1.0,
582
+ "f":1.0
583
+ },
584
+ "PartType":{
585
+ "p":1.0,
586
+ "r":0.9945652174,
587
+ "f":0.9972752044
588
+ },
589
+ "VerbType":{
590
+ "p":0.9820895522,
591
+ "r":0.9820895522,
592
+ "f":0.9820895522
593
+ },
594
+ "Poss":{
595
+ "p":0.9242424242,
596
+ "r":1.0,
597
+ "f":0.9606299213
598
+ },
599
+ "Reflex":{
600
+ "p":1.0,
601
+ "r":0.75,
602
+ "f":0.8571428571
603
+ }
604
+ },
605
+ "dep_uas":0.9078427294,
606
+ "dep_las":0.8749739963,
607
+ "dep_las_per_type":{
608
+ "amod":{
609
+ "p":0.7847082495,
610
+ "r":0.8423326134,
611
+ "f":0.8125
612
+ },
613
+ "nsubj":{
614
+ "p":0.8905775076,
615
+ "r":0.8694362018,
616
+ "f":0.8798798799
617
+ },
618
+ "root":{
619
+ "p":0.9381443299,
620
+ "r":0.9381443299,
621
+ "f":0.9381443299
622
+ },
623
+ "det":{
624
+ "p":0.9713831479,
625
+ "r":0.9546875,
626
+ "f":0.962962963
627
+ },
628
+ "obj":{
629
+ "p":0.8823529412,
630
+ "r":0.8585858586,
631
+ "f":0.8703071672
632
+ },
633
+ "case":{
634
+ "p":0.9578783151,
635
+ "r":0.9684542587,
636
+ "f":0.9631372549
637
+ },
638
+ "nmod":{
639
+ "p":0.8211678832,
640
+ "r":0.8272058824,
641
+ "f":0.8241758242
642
+ },
643
+ "mark":{
644
+ "p":0.9285714286,
645
+ "r":0.8888888889,
646
+ "f":0.9082969432
647
+ },
648
+ "xcomp":{
649
+ "p":0.7714285714,
650
+ "r":0.675,
651
+ "f":0.72
652
+ },
653
+ "aux":{
654
+ "p":0.9397590361,
655
+ "r":0.9285714286,
656
+ "f":0.9341317365
657
+ },
658
+ "dep":{
659
+ "p":0.7515151515,
660
+ "r":0.5961538462,
661
+ "f":0.6648793566
662
+ },
663
+ "cc":{
664
+ "p":0.8552631579,
665
+ "r":0.8705357143,
666
+ "f":0.8628318584
667
+ },
668
+ "conj":{
669
+ "p":0.8486055777,
670
+ "r":0.8486055777,
671
+ "f":0.8486055777
672
+ },
673
+ "obl":{
674
+ "p":0.8155619597,
675
+ "r":0.8299120235,
676
+ "f":0.8226744186
677
+ },
678
+ "cop":{
679
+ "p":0.8970588235,
680
+ "r":0.8133333333,
681
+ "f":0.8531468531
682
+ },
683
+ "ccomp":{
684
+ "p":0.8514851485,
685
+ "r":0.8686868687,
686
+ "f":0.86
687
+ },
688
+ "advmod":{
689
+ "p":0.7912621359,
690
+ "r":0.8232323232,
691
+ "f":0.8069306931
692
+ },
693
+ "compound:prt":{
694
+ "p":0.9705882353,
695
+ "r":0.9428571429,
696
+ "f":0.9565217391
697
+ },
698
+ "aux:pass":{
699
+ "p":0.9782608696,
700
+ "r":1.0,
701
+ "f":0.989010989
702
+ },
703
+ "nummod":{
704
+ "p":0.7073170732,
705
+ "r":0.90625,
706
+ "f":0.7945205479
707
+ },
708
+ "nsubj:pass":{
709
+ "p":0.6724137931,
710
+ "r":0.7959183673,
711
+ "f":0.7289719626
712
+ },
713
+ "iobj":{
714
+ "p":1.0,
715
+ "r":0.625,
716
+ "f":0.7692307692
717
+ },
718
+ "flat":{
719
+ "p":0.6756756757,
720
+ "r":0.8333333333,
721
+ "f":0.7462686567
722
+ },
723
+ "appos":{
724
+ "p":0.4285714286,
725
+ "r":0.5,
726
+ "f":0.4615384615
727
+ }
728
+ },
729
+ "lemma_acc":0.9786593964
730
+ },
731
+ "requirements":[
732
+ "spacy-experimental>=0.2.0,<0.3.0",
733
+ "spacy-transformers>=1.1.3,<1.2.0"
734
+ ]
735
+ }
AF/af_udv25_afrikaansafribooms_trf/morphologizer/cfg ADDED
@@ -0,0 +1,130 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "extend":false,
3
+ "labels_morph":{
4
+ "Definite=Def|POS=DET|PronType=Art":"Definite=Def|PronType=Art",
5
+ "Number=Sing|POS=NOUN":"Number=Sing",
6
+ "AdpType=Prep|POS=ADP":"AdpType=Prep",
7
+ "AdjType=Attr|Case=Nom|Degree=Pos|POS=ADJ":"AdjType=Attr|Case=Nom|Degree=Pos",
8
+ "Number=Plur|POS=NOUN":"Number=Plur",
9
+ "POS=AUX|Tense=Pres|VerbForm=Fin,Inf|VerbType=Cop":"Tense=Pres|VerbForm=Fin,Inf|VerbType=Cop",
10
+ "Definite=Ind|POS=DET|PronType=Art":"Definite=Ind|PronType=Art",
11
+ "POS=NUM":"",
12
+ "POS=PART|PartType=Inf":"PartType=Inf",
13
+ "POS=VERB|Subcat=Tran|Tense=Pres|VerbForm=Fin,Inf":"Subcat=Tran|Tense=Pres|VerbForm=Fin,Inf",
14
+ "POS=PRON|PronType=Rel":"PronType=Rel",
15
+ "POS=AUX|Tense=Pres|VerbForm=Fin,Inf|VerbType=Pas":"Tense=Pres|VerbForm=Fin,Inf|VerbType=Pas",
16
+ "POS=PUNCT":"",
17
+ "POS=CCONJ":"",
18
+ "POS=SCONJ":"",
19
+ "POS=VERB|Subcat=Intr|Tense=Pres|VerbForm=Fin,Inf":"Subcat=Intr|Tense=Pres|VerbForm=Fin,Inf",
20
+ "POS=VERB|Subcat=Intr|Tense=Past|VerbForm=Part":"Subcat=Intr|Tense=Past|VerbForm=Part",
21
+ "POS=AUX|Tense=Past|VerbForm=Fin|VerbType=Pas":"Tense=Past|VerbForm=Fin|VerbType=Pas",
22
+ "Degree=Pos|POS=ADV":"Degree=Pos",
23
+ "POS=AUX|Tense=Pres|VerbForm=Fin,Inf|VerbType=Mod":"Tense=Pres|VerbForm=Fin,Inf|VerbType=Mod",
24
+ "POS=DET|PronType=Ind":"PronType=Ind",
25
+ "POS=X":"",
26
+ "Number=Sing|POS=PROPN":"Number=Sing",
27
+ "POS=PRON|PronType=Ind":"PronType=Ind",
28
+ "POS=PART|PartType=Neg":"PartType=Neg",
29
+ "POS=VERB|Subcat=Tran|Tense=Past|VerbForm=Part":"Subcat=Tran|Tense=Past|VerbForm=Part",
30
+ "AdjType=Pred|Case=Nom|Degree=Pos|POS=ADJ":"AdjType=Pred|Case=Nom|Degree=Pos",
31
+ "POS=DET|PronType=Dem":"PronType=Dem",
32
+ "Degree=Cmp|POS=ADV":"Degree=Cmp",
33
+ "Case=Nom|Number=Sing|POS=PRON|Person=3|PronType=Prs":"Case=Nom|Number=Sing|Person=3|PronType=Prs",
34
+ "POS=SYM":"",
35
+ "Case=Acc,Nom|Number=Plur|POS=PRON|Person=1|PronType=Prs":"Case=Acc,Nom|Number=Plur|Person=1|PronType=Prs",
36
+ "POS=PART|PartType=Gen":"PartType=Gen",
37
+ "Case=Acc|Number=Sing|POS=PRON|Person=2|PronType=Prs":"Case=Acc|Number=Sing|Person=2|PronType=Prs",
38
+ "Number=Sing|POS=PRON|Person=2|PronType=Prs|Reflex=Yes":"Number=Sing|Person=2|PronType=Prs|Reflex=Yes",
39
+ "Degree=Sup|POS=ADV":"Degree=Sup",
40
+ "Degree=Dim|Number=Sing|POS=NOUN":"Degree=Dim|Number=Sing",
41
+ "Number=Sing|POS=PRON|Person=2|Poss=Yes|PronType=Prs":"Number=Sing|Person=2|Poss=Yes|PronType=Prs",
42
+ "POS=PRON|PronType=Int":"PronType=Int",
43
+ "Number=Plur|POS=PRON|Person=1|Poss=Yes|PronType=Prs":"Number=Plur|Person=1|Poss=Yes|PronType=Prs",
44
+ "Number=Sing|POS=PRON|Person=3|PronType=Prs|Reflex=Yes":"Number=Sing|Person=3|PronType=Prs|Reflex=Yes",
45
+ "Number=Plur|POS=PRON|Person=3|Poss=Yes|PronType=Prs":"Number=Plur|Person=3|Poss=Yes|PronType=Prs",
46
+ "AdjType=Attr|Case=Nom|Degree=Sup|POS=ADJ":"AdjType=Attr|Case=Nom|Degree=Sup",
47
+ "Case=Nom|Number=Sing|POS=PRON|Person=2|PronType=Prs":"Case=Nom|Number=Sing|Person=2|PronType=Prs",
48
+ "AdjType=Pred|Case=Nom|Degree=Cmp|POS=ADJ":"AdjType=Pred|Case=Nom|Degree=Cmp",
49
+ "POS=VERB|Subcat=Prep|Tense=Pres|VerbForm=Fin,Inf":"Subcat=Prep|Tense=Pres|VerbForm=Fin,Inf",
50
+ "POS=AUX|Tense=Pres|VerbForm=Fin,Inf|VerbType=Aux":"Tense=Pres|VerbForm=Fin,Inf|VerbType=Aux",
51
+ "Number=Sing|POS=PRON|Person=3|Poss=Yes|PronType=Prs":"Number=Sing|Person=3|Poss=Yes|PronType=Prs",
52
+ "POS=PRON|PronType=Rcp":"PronType=Rcp",
53
+ "POS=AUX|Tense=Past|VerbForm=Fin|VerbType=Mod":"Tense=Past|VerbForm=Fin|VerbType=Mod",
54
+ "Case=Acc,Nom|Number=Plur|POS=PRON|Person=3|PronType=Prs":"Case=Acc,Nom|Number=Plur|Person=3|PronType=Prs",
55
+ "POS=AUX|Tense=Past|VerbForm=Fin|VerbType=Cop":"Tense=Past|VerbForm=Fin|VerbType=Cop",
56
+ "Case=Acc|Number=Sing|POS=PRON|Person=3|PronType=Prs":"Case=Acc|Number=Sing|Person=3|PronType=Prs",
57
+ "Case=Nom|Number=Sing|POS=PRON|Person=1|PronType=Prs":"Case=Nom|Number=Sing|Person=1|PronType=Prs",
58
+ "Number=Sing|POS=PRON|Person=1|Poss=Yes|PronType=Prs":"Number=Sing|Person=1|Poss=Yes|PronType=Prs",
59
+ "Case=Acc,Nom|Number=Plur|POS=PRON|Person=2|PronType=Prs":"Case=Acc,Nom|Number=Plur|Person=2|PronType=Prs",
60
+ "Number=Plur|POS=PRON|Person=3|PronType=Prs|Reflex=Yes":"Number=Plur|Person=3|PronType=Prs|Reflex=Yes",
61
+ "AdjType=Attr|Case=Nom|Degree=Cmp|POS=ADJ":"AdjType=Attr|Case=Nom|Degree=Cmp",
62
+ "Number=Plur|POS=PRON|Person=1|PronType=Prs|Reflex=Yes":"Number=Plur|Person=1|PronType=Prs|Reflex=Yes",
63
+ "Case=Acc|Number=Sing|POS=PRON|Person=1|PronType=Prs":"Case=Acc|Number=Sing|Person=1|PronType=Prs",
64
+ "AdjType=Pred|Case=Nom|Degree=Sup|POS=ADJ":"AdjType=Pred|Case=Nom|Degree=Sup"
65
+ },
66
+ "labels_pos":{
67
+ "Definite=Def|POS=DET|PronType=Art":90,
68
+ "Number=Sing|POS=NOUN":92,
69
+ "AdpType=Prep|POS=ADP":85,
70
+ "AdjType=Attr|Case=Nom|Degree=Pos|POS=ADJ":84,
71
+ "Number=Plur|POS=NOUN":92,
72
+ "POS=AUX|Tense=Pres|VerbForm=Fin,Inf|VerbType=Cop":87,
73
+ "Definite=Ind|POS=DET|PronType=Art":90,
74
+ "POS=NUM":93,
75
+ "POS=PART|PartType=Inf":94,
76
+ "POS=VERB|Subcat=Tran|Tense=Pres|VerbForm=Fin,Inf":100,
77
+ "POS=PRON|PronType=Rel":95,
78
+ "POS=AUX|Tense=Pres|VerbForm=Fin,Inf|VerbType=Pas":87,
79
+ "POS=PUNCT":97,
80
+ "POS=CCONJ":89,
81
+ "POS=SCONJ":98,
82
+ "POS=VERB|Subcat=Intr|Tense=Pres|VerbForm=Fin,Inf":100,
83
+ "POS=VERB|Subcat=Intr|Tense=Past|VerbForm=Part":100,
84
+ "POS=AUX|Tense=Past|VerbForm=Fin|VerbType=Pas":87,
85
+ "Degree=Pos|POS=ADV":86,
86
+ "POS=AUX|Tense=Pres|VerbForm=Fin,Inf|VerbType=Mod":87,
87
+ "POS=DET|PronType=Ind":90,
88
+ "POS=X":101,
89
+ "Number=Sing|POS=PROPN":96,
90
+ "POS=PRON|PronType=Ind":95,
91
+ "POS=PART|PartType=Neg":94,
92
+ "POS=VERB|Subcat=Tran|Tense=Past|VerbForm=Part":100,
93
+ "AdjType=Pred|Case=Nom|Degree=Pos|POS=ADJ":84,
94
+ "POS=DET|PronType=Dem":90,
95
+ "Degree=Cmp|POS=ADV":86,
96
+ "Case=Nom|Number=Sing|POS=PRON|Person=3|PronType=Prs":95,
97
+ "POS=SYM":99,
98
+ "Case=Acc,Nom|Number=Plur|POS=PRON|Person=1|PronType=Prs":95,
99
+ "POS=PART|PartType=Gen":94,
100
+ "Case=Acc|Number=Sing|POS=PRON|Person=2|PronType=Prs":95,
101
+ "Number=Sing|POS=PRON|Person=2|PronType=Prs|Reflex=Yes":95,
102
+ "Degree=Sup|POS=ADV":86,
103
+ "Degree=Dim|Number=Sing|POS=NOUN":92,
104
+ "Number=Sing|POS=PRON|Person=2|Poss=Yes|PronType=Prs":95,
105
+ "POS=PRON|PronType=Int":95,
106
+ "Number=Plur|POS=PRON|Person=1|Poss=Yes|PronType=Prs":95,
107
+ "Number=Sing|POS=PRON|Person=3|PronType=Prs|Reflex=Yes":95,
108
+ "Number=Plur|POS=PRON|Person=3|Poss=Yes|PronType=Prs":95,
109
+ "AdjType=Attr|Case=Nom|Degree=Sup|POS=ADJ":84,
110
+ "Case=Nom|Number=Sing|POS=PRON|Person=2|PronType=Prs":95,
111
+ "AdjType=Pred|Case=Nom|Degree=Cmp|POS=ADJ":84,
112
+ "POS=VERB|Subcat=Prep|Tense=Pres|VerbForm=Fin,Inf":100,
113
+ "POS=AUX|Tense=Pres|VerbForm=Fin,Inf|VerbType=Aux":87,
114
+ "Number=Sing|POS=PRON|Person=3|Poss=Yes|PronType=Prs":95,
115
+ "POS=PRON|PronType=Rcp":95,
116
+ "POS=AUX|Tense=Past|VerbForm=Fin|VerbType=Mod":87,
117
+ "Case=Acc,Nom|Number=Plur|POS=PRON|Person=3|PronType=Prs":95,
118
+ "POS=AUX|Tense=Past|VerbForm=Fin|VerbType=Cop":87,
119
+ "Case=Acc|Number=Sing|POS=PRON|Person=3|PronType=Prs":95,
120
+ "Case=Nom|Number=Sing|POS=PRON|Person=1|PronType=Prs":95,
121
+ "Number=Sing|POS=PRON|Person=1|Poss=Yes|PronType=Prs":95,
122
+ "Case=Acc,Nom|Number=Plur|POS=PRON|Person=2|PronType=Prs":95,
123
+ "Number=Plur|POS=PRON|Person=3|PronType=Prs|Reflex=Yes":95,
124
+ "AdjType=Attr|Case=Nom|Degree=Cmp|POS=ADJ":84,
125
+ "Number=Plur|POS=PRON|Person=1|PronType=Prs|Reflex=Yes":95,
126
+ "Case=Acc|Number=Sing|POS=PRON|Person=1|PronType=Prs":95,
127
+ "AdjType=Pred|Case=Nom|Degree=Sup|POS=ADJ":84
128
+ },
129
+ "overwrite":false
130
+ }
AF/af_udv25_afrikaansafribooms_trf/morphologizer/model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16dda1c66e95145dbb8553c64200c8540d82d37c8cb99ab4764eb734f0c55069
3
+ size 188226
AF/af_udv25_afrikaansafribooms_trf/parser/cfg ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "moves":null,
3
+ "update_with_oracle_cut_size":100,
4
+ "multitasks":[
5
+
6
+ ],
7
+ "min_action_freq":5,
8
+ "learn_tokens":false,
9
+ "beam_width":1,
10
+ "beam_density":0.0,
11
+ "beam_update_prob":0.0,
12
+ "incorrect_spans_key":null
13
+ }
AF/af_udv25_afrikaansafribooms_trf/parser/model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:844d75edee6630df2c2d324fc146f8f9b7ee95dc6d77829f29b62bb6ccece85d
3
+ size 591617
AF/af_udv25_afrikaansafribooms_trf/parser/moves ADDED
@@ -0,0 +1 @@
 
 
1
+ ��moves�9{"0":{"":18776},"1":{"":13805},"2":{"case":4223,"det":4041,"amod":2134,"nsubj":1929,"obj":1368,"obl":1060,"aux":803,"advmod":673,"mark":629,"nsubj:pass":393,"nummod":327,"punct":244,"cop":224,"nmod":174,"compound:prt":144,"aux:pass":102,"advmod||cop":79,"dep":58,"aux||cop":40,"iobj":22,"cc":20,"ccomp":18,"amod||cop":15,"xcomp":14,"case||cop":10,"obl||cop":5,"nsubj||cop":5},"3":{"punct":2775,"nmod":1926,"conj":1440,"cc":1201,"dep":1066,"obl":1043,"amod":819,"ccomp":623,"aux:pass":565,"obj":468,"advmod":346,"cop":272,"aux":227,"xcomp":177,"flat":145,"case":103,"compound:prt":96,"punct||cop":82,"obl||case":73,"cc||case":71,"appos":47,"nsubj":45,"mark":36,"amod||cop":32,"iobj":20,"advmod||cop":15,"punct||case":12,"cc||nsubj":10,"ccomp||nsubj":10,"conj||cop":9,"cc||cop":8,"mark||cop":7,"cop||nmod":5},"4":{"ROOT":1315}}�cfg��neg_key�
AF/af_udv25_afrikaansafribooms_trf/senter/cfg ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "overwrite":false
3
+ }
AF/af_udv25_afrikaansafribooms_trf/senter/model ADDED
Binary file (6.74 kB). View file
 
AF/af_udv25_afrikaansafribooms_trf/tagger/cfg ADDED
@@ -0,0 +1,101 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "labels":[
3
+ "AOA",
4
+ "AOP",
5
+ "ASA",
6
+ "ASP",
7
+ "AVA",
8
+ "AVP",
9
+ "BO",
10
+ "BS",
11
+ "BV",
12
+ "KN",
13
+ "KO",
14
+ "LB",
15
+ "LO",
16
+ "NA",
17
+ "NEE",
18
+ "NM",
19
+ "NME",
20
+ "NSE",
21
+ "NSED",
22
+ "NSM",
23
+ "PA",
24
+ "PB",
25
+ "PDHEB",
26
+ "PDHEDP",
27
+ "PDHENP",
28
+ "PDHEW",
29
+ "PDMB",
30
+ "PDMP",
31
+ "PDMW",
32
+ "PDOENP",
33
+ "PDOEW",
34
+ "PDVEB",
35
+ "PDVEDP",
36
+ "PDVENP",
37
+ "PDVEW",
38
+ "PEEB",
39
+ "PEEDP",
40
+ "PEENP",
41
+ "PEMB",
42
+ "PEMP",
43
+ "PEMW",
44
+ "PO",
45
+ "PTEB",
46
+ "PTEDP",
47
+ "PTENP",
48
+ "PTEW",
49
+ "PTMP",
50
+ "PV",
51
+ "PW",
52
+ "RA",
53
+ "RK",
54
+ "RL",
55
+ "RO",
56
+ "RS",
57
+ "RSF",
58
+ "RV",
59
+ "RWD",
60
+ "SVS",
61
+ "THAB",
62
+ "THAO",
63
+ "THBB",
64
+ "THBO",
65
+ "THNB",
66
+ "THPB",
67
+ "THPO",
68
+ "TRAB",
69
+ "TRAO",
70
+ "TRBB",
71
+ "UPB",
72
+ "UPD",
73
+ "UPI",
74
+ "UPO",
75
+ "UPS",
76
+ "UPV",
77
+ "UPW",
78
+ "UXD",
79
+ "VTHOG",
80
+ "VTHOK",
81
+ "VTHOO",
82
+ "VTHOV",
83
+ "VTHSG",
84
+ "VTHSO",
85
+ "VTUOA",
86
+ "VTUOM",
87
+ "VTUOP",
88
+ "VUOT",
89
+ "VVHOG",
90
+ "VVHOK",
91
+ "VVHOO",
92
+ "VVUOM",
93
+ "VVUOP",
94
+ "ZE",
95
+ "ZM",
96
+ "ZPL",
97
+ "ZPR"
98
+ ],
99
+ "neg_prefix":"!",
100
+ "overwrite":false
101
+ }
AF/af_udv25_afrikaansafribooms_trf/tagger/model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e09c3358834624e8c33485b95a8204cf524ca8f8d8d07d72b14c762e7c18bfb0
3
+ size 292811
AF/af_udv25_afrikaansafribooms_trf/transformer/cfg ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "max_batch_items":4096
3
+ }
AF/af_udv25_afrikaansafribooms_trf/transformer/model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0a65559556652a5714561cf861942d697a8ebcc5e672e32c49d2c6876e45437
3
+ size 1126406104
AF/af_udv25_afrikaansafribooms_trf/vocab/key2row ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76be8b528d0075f7aae98d6fa57a6d3c83ae480a8469e668d7b0af968995ac71
3
+ size 1
AF/af_udv25_afrikaansafribooms_trf/vocab/lookups.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76be8b528d0075f7aae98d6fa57a6d3c83ae480a8469e668d7b0af968995ac71
3
+ size 1
AF/af_udv25_afrikaansafribooms_trf/vocab/strings.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8719f892044d9cd123d46078dd8c1b13dcc09d7ca38d122f71bb4b47e4650e35
3
+ size 275533
AF/af_udv25_afrikaansafribooms_trf/vocab/vectors ADDED
Binary file (128 Bytes). View file
 
AF/af_udv25_afrikaansafribooms_trf/vocab/vectors.cfg ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "mode":"default"
3
+ }
EL/el_cop_sigtyp_trf/.gitattributes ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz filter=lfs diff=lfs merge=lfs -text
33
+ *.zip filter=lfs diff=lfs merge=lfs -text
34
+ *.zst filter=lfs diff=lfs merge=lfs -text
35
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ el_cop_sigtyp_trf-any-py3-none-any.whl filter=lfs diff=lfs merge=lfs -text
37
+ transformer/model filter=lfs diff=lfs merge=lfs -text
38
+ morphologizer/model filter=lfs diff=lfs merge=lfs -text
39
+ tagger/model filter=lfs diff=lfs merge=lfs -text
40
+ trainable_lemmatizer/model filter=lfs diff=lfs merge=lfs -text
EL/el_cop_sigtyp_trf/README.md ADDED
The diff for this file is too large to render. See raw diff
 
EL/el_cop_sigtyp_trf/config.cfg ADDED
@@ -0,0 +1,211 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [paths]
2
+ train = "corpus/train/cop_train.spacy"
3
+ dev = "corpus/dev/cop_valid.spacy"
4
+ vectors = null
5
+ init_tok2vec = null
6
+
7
+ [system]
8
+ gpu_allocator = "pytorch"
9
+ seed = 42
10
+
11
+ [nlp]
12
+ lang = "el"
13
+ pipeline = ["transformer","parser","trainable_lemmatizer","tagger","morphologizer"]
14
+ batch_size = 128
15
+ disabled = []
16
+ before_creation = null
17
+ after_creation = null
18
+ after_pipeline_creation = null
19
+ tokenizer = {"@tokenizers":"spacy.Tokenizer.v1"}
20
+
21
+ [components]
22
+
23
+ [components.morphologizer]
24
+ factory = "morphologizer"
25
+ extend = false
26
+ label_smoothing = 0.0
27
+ overwrite = true
28
+ scorer = {"@scorers":"spacy.morphologizer_scorer.v1"}
29
+
30
+ [components.morphologizer.model]
31
+ @architectures = "spacy.Tagger.v2"
32
+ nO = null
33
+ normalize = false
34
+
35
+ [components.morphologizer.model.tok2vec]
36
+ @architectures = "spacy-transformers.TransformerListener.v1"
37
+ grad_factor = 1.0
38
+ pooling = {"@layers":"reduce_mean.v1"}
39
+ upstream = "*"
40
+
41
+ [components.parser]
42
+ factory = "parser"
43
+ learn_tokens = false
44
+ min_action_freq = 30
45
+ moves = null
46
+ scorer = {"@scorers":"spacy.parser_scorer.v1"}
47
+ update_with_oracle_cut_size = 100
48
+
49
+ [components.parser.model]
50
+ @architectures = "spacy.TransitionBasedParser.v2"
51
+ state_type = "parser"
52
+ extra_state_tokens = false
53
+ hidden_width = 128
54
+ maxout_pieces = 3
55
+ use_upper = false
56
+ nO = null
57
+
58
+ [components.parser.model.tok2vec]
59
+ @architectures = "spacy-transformers.TransformerListener.v1"
60
+ grad_factor = 1.0
61
+ pooling = {"@layers":"reduce_mean.v1"}
62
+ upstream = "*"
63
+
64
+ [components.tagger]
65
+ factory = "tagger"
66
+ label_smoothing = 0.0
67
+ neg_prefix = "!"
68
+ overwrite = false
69
+ scorer = {"@scorers":"spacy.tagger_scorer.v1"}
70
+
71
+ [components.tagger.model]
72
+ @architectures = "spacy.Tagger.v2"
73
+ nO = null
74
+ normalize = false
75
+
76
+ [components.tagger.model.tok2vec]
77
+ @architectures = "spacy-transformers.TransformerListener.v1"
78
+ grad_factor = 1.0
79
+ pooling = {"@layers":"reduce_mean.v1"}
80
+ upstream = "*"
81
+
82
+ [components.trainable_lemmatizer]
83
+ factory = "trainable_lemmatizer"
84
+ backoff = "orth"
85
+ min_tree_freq = 3
86
+ overwrite = false
87
+ scorer = {"@scorers":"spacy.lemmatizer_scorer.v1"}
88
+ top_k = 1
89
+
90
+ [components.trainable_lemmatizer.model]
91
+ @architectures = "spacy.Tagger.v2"
92
+ nO = null
93
+ normalize = false
94
+
95
+ [components.trainable_lemmatizer.model.tok2vec]
96
+ @architectures = "spacy-transformers.TransformerListener.v1"
97
+ grad_factor = 1.0
98
+ pooling = {"@layers":"reduce_mean.v1"}
99
+ upstream = "*"
100
+
101
+ [components.transformer]
102
+ factory = "transformer"
103
+ max_batch_items = 4096
104
+ set_extra_annotations = {"@annotation_setters":"spacy-transformers.null_annotation_setter.v1"}
105
+
106
+ [components.transformer.model]
107
+ @architectures = "spacy-transformers.TransformerModel.v3"
108
+ name = "ljvmiranda921/LiBERTus-base"
109
+ mixed_precision = false
110
+
111
+ [components.transformer.model.get_spans]
112
+ @span_getters = "spacy-transformers.strided_spans.v1"
113
+ window = 128
114
+ stride = 96
115
+
116
+ [components.transformer.model.grad_scaler_config]
117
+
118
+ [components.transformer.model.tokenizer_config]
119
+ use_fast = true
120
+
121
+ [components.transformer.model.transformer_config]
122
+
123
+ [corpora]
124
+
125
+ [corpora.dev]
126
+ @readers = "spacy.Corpus.v1"
127
+ path = ${paths.dev}
128
+ max_length = 0
129
+ gold_preproc = false
130
+ limit = 0
131
+ augmenter = null
132
+
133
+ [corpora.train]
134
+ @readers = "spacy.Corpus.v1"
135
+ path = ${paths.train}
136
+ max_length = 0
137
+ gold_preproc = false
138
+ limit = 0
139
+ augmenter = null
140
+
141
+ [training]
142
+ accumulate_gradient = 3
143
+ dev_corpus = "corpora.dev"
144
+ train_corpus = "corpora.train"
145
+ seed = ${system.seed}
146
+ gpu_allocator = ${system.gpu_allocator}
147
+ dropout = 0.1
148
+ patience = 1600
149
+ max_epochs = 0
150
+ max_steps = 20000
151
+ eval_frequency = 200
152
+ frozen_components = []
153
+ annotating_components = []
154
+ before_to_disk = null
155
+ before_update = null
156
+
157
+ [training.batcher]
158
+ @batchers = "spacy.batch_by_padded.v1"
159
+ discard_oversize = true
160
+ size = 2000
161
+ buffer = 256
162
+ get_length = null
163
+
164
+ [training.logger]
165
+ @loggers = "spacy.ConsoleLogger.v3"
166
+ console_output = true
167
+ progress_bar = "eval"
168
+ output_file = null
169
+
170
+ [training.optimizer]
171
+ @optimizers = "Adam.v1"
172
+ beta1 = 0.9
173
+ beta2 = 0.999
174
+ L2_is_weight_decay = true
175
+ L2 = 0.01
176
+ grad_clip = 1.0
177
+ use_averages = false
178
+ eps = 0.00000001
179
+
180
+ [training.optimizer.learn_rate]
181
+ @schedules = "warmup_linear.v1"
182
+ warmup_steps = 250
183
+ total_steps = 20000
184
+ initial_rate = 0.00005
185
+
186
+ [training.score_weights]
187
+ dep_uas = 0.09
188
+ dep_las = 0.09
189
+ dep_las_per_type = null
190
+ sents_p = null
191
+ sents_r = null
192
+ sents_f = 0.0
193
+ lemma_acc = 0.17
194
+ tag_acc = 0.17
195
+ pos_acc = 0.25
196
+ morph_acc = 0.25
197
+ morph_per_feat = null
198
+
199
+ [pretraining]
200
+
201
+ [initialize]
202
+ vectors = ${paths.vectors}
203
+ init_tok2vec = ${paths.init_tok2vec}
204
+ vocab_data = null
205
+ lookups = null
206
+ before_init = null
207
+ after_init = null
208
+
209
+ [initialize.components]
210
+
211
+ [initialize.tokenizer]
EL/el_cop_sigtyp_trf/el_cop_sigtyp_trf-any-py3-none-any.whl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a59340bb99ddc6cd3989fd0f657d422bbae7ccd44883ec98433346a85b2b2ca
3
+ size 477444124
EL/el_cop_sigtyp_trf/meta.json ADDED
The diff for this file is too large to render. See raw diff
 
EL/el_cop_sigtyp_trf/morphologizer/cfg ADDED
The diff for this file is too large to render. See raw diff
 
EL/el_cop_sigtyp_trf/morphologizer/model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:537e4f4befa9bda11c850fbaaec23ea5b161f56241aba1d0833e76eebf95a82f
3
+ size 2153853
EL/el_cop_sigtyp_trf/parser/cfg ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "moves":null,
3
+ "update_with_oracle_cut_size":100,
4
+ "multitasks":[
5
+
6
+ ],
7
+ "min_action_freq":30,
8
+ "learn_tokens":false,
9
+ "beam_width":1,
10
+ "beam_density":0.0,
11
+ "beam_update_prob":0.0,
12
+ "incorrect_spans_key":null
13
+ }
EL/el_cop_sigtyp_trf/parser/model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6a656fb7d6d87a95cc97246b9f0dca7fddd16f6c7c9c9c80a5cfbe5182ca8a8
3
+ size 853263
EL/el_cop_sigtyp_trf/parser/moves ADDED
@@ -0,0 +1 @@
 
 
1
+ ��moves��{"0":{"":4818},"1":{"":14316},"2":{"cc":950,"punct":841,"advmod":635,"mark":586,"dislocated":426,"advcl":298,"case":271,"aux":219,"obl":155,"nsubj":97,"discourse":46,"vocative":30,"dep":0},"3":{"obl":3413,"punct":1924,"nmod":1407,"advmod":1333,"conj":952,"advcl":864,"obj":834,"acl:relcl":664,"ccomp":574,"parataxis":534,"cop":368,"appos":312,"xcomp":304,"dislocated":194,"nsubj":156,"det":107,"vocative":79,"obl:npmod":52,"csubj":45,"flat":40,"dep":0},"4":{"ROOT":1730}}�cfg��neg_key�
EL/el_cop_sigtyp_trf/tagger/cfg ADDED
The diff for this file is too large to render. See raw diff
 
EL/el_cop_sigtyp_trf/tagger/model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:312de10ac7419abc3cf02141eb8b5c58870e9e444c814142c344e97e6ed59a6f
3
+ size 6392581
EL/el_cop_sigtyp_trf/tokenizer ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ ��prefix_search� �^''|^§|^%|^=|^\+[0-9]+%|^\'([0-9]){2}([\-]\'([0-9]){2})*|^\-([0-9]){1,9}\.([0-9]){1,9}|^\'([Α-Ωα-ωίϊΐόάέύϋΰήώ]+)\'|^([Α-Ωα-ωίϊΐόάέύϋΰήώ]){1,3}\'|^http://www.[A-Za-z]+\-[A-Za-z]+(\.[A-Za-z]+)+(\/[A-Za-z]+)*(\.[A-Za-z]+)*|^[ΈΆΊΑ-Ωα-ωίϊΐόάέύϋΰήώ]+\*|^\$([0-9])+([\,\.]([0-9])+){0,1}|^…|^……|^,|^:|^;|^\!|^\?|^¿|^؟|^¡|^\(|^\)|^\[|^\]|^\{|^\}|^<|^>|^_|^#|^\*|^&|^。|^？|^！|^，|^、|^；|^：|^～|^·|^।|^،|^۔|^؛|^٪|^\.\.+|^…|^\'|^"|^”|^“|^`|^‘|^´|^’|^‚|^,|^„|^»|^«|^「|^」|^『|^』|^（|^）|^〔|^〕|^【|^】|^《|^》|^〈|^〉|^〈|^〉|^⟦|^⟧|^\$|^£|^€|^¥|^฿|^US\$|^C\$|^A\$|^₽|^﷼|^₴|^₠|^₡|^₢|^₣|^₤|^₥|^₦|^₧|^₨|^₩|^₪|^₫|^€|^₭|^₮|^₯|^₰|^₱|^₲|^₳|^₴|^₵|^₶|^₷|^₸|^₹|^₺|^₻|^₼|^₽|^₾|^₿|^[\u00A6\u00A9\u00AE\u00B0\u0482\u058D\u058E\u060E\u060F\u06DE\u06E9\u06FD\u06FE\u07F6\u09FA\u0B70\u0BF3-\u0BF8\u0BFA\u0C7F\u0D4F\u0D79\u0F01-\u0F03\u0F13\u0F15-\u0F17\u0F1A-\u0F1F\u0F34\u0F36\u0F38\u0FBE-\u0FC5\u0FC7-\u0FCC\u0FCE\u0FCF\u0FD5-\u0FD8\u109E\u109F\u1390-\u1399\u1940\u19DE-\u19FF\u1B61-\u1B6A\u1B74-\u1B7C\u2100\u2101\u2103-\u2106\u2108\u2109\u2114\u2116\u2117\u211E-\u2123\u2125\u2127\u2129\u212E\u213A\u213B\u214A\u214C\u214D\u214F\u218A\u218B\u2195-\u2199\u219C-\u219F\u21A1\u21A2\u21A4\u21A5\u21A7-\u21AD\u21AF-\u21CD\u21D0\u21D1\u21D3\u21D5-\u21F3\u2300-\u2307\u230C-\u231F\u2322-\u2328\u232B-\u237B\u237D-\u239A\u23B4-\u23DB\u23E2-\u2426\u2440-\u244A\u249C-\u24E9\u2500-\u25B6\u25B8-\u25C0\u25C2-\u25F7\u2600-\u266E\u2670-\u2767\u2794-\u27BF\u2800-\u28FF\u2B00-\u2B2F\u2B45\u2B46\u2B4D-\u2B73\u2B76-\u2B95\u2B98-\u2BC8\u2BCA-\u2BFE\u2CE5-\u2CEA\u2E80-\u2E99\u2E9B-\u2EF3\u2F00-\u2FD5\u2FF0-\u2FFB\u3004\u3012\u3013\u3020\u3036\u3037\u303E\u303F\u3190\u3191\u3196-\u319F\u31C0-\u31E3\u3200-\u321E\u322A-\u3247\u3250\u3260-\u327F\u328A-\u32B0\u32C0-\u32FE\u3300-\u33FF\u4DC0-\u4DFF\uA490-\uA4C6\uA828-\uA82B\uA836\uA837\uA839\uAA77-\uAA79\uFDFD\uFFE4\uFFE8\uFFED\uFFEE\uFFFC\uFFFD\U00010137-\U0001013F\U00010179-\U00010189\U0001018C-\U0001018E\U00010190-\U0001019B\U000101A0\U000101D0-\U000101FC\U00010877\U00010878\U00010AC8\U0001173F\U00016B3C-\U00016B3F\U00016B45\U0001BC9C\U0001D000-\U0001D0F5\U0001D100-\U0001D126\U0001D129-\U0001D164\U0001D16A-\U0001D16C\U0001D183\U0001D184\U0001D18C-\U0001D1A9\U0001D1AE-\U0001D1E8\U0001D200-\U0001D241\U0001D245\U0001D300-\U0001D356\U0001D800-\U0001D9FF\U0001DA37-\U0001DA3A\U0001DA6D-\U0001DA74\U0001DA76-\U0001DA83\U0001DA85\U0001DA86\U0001ECAC\U0001F000-\U0001F02B\U0001F030-\U0001F093\U0001F0A0-\U0001F0AE\U0001F0B1-\U0001F0BF\U0001F0C1-\U0001F0CF\U0001F0D1-\U0001F0F5\U0001F110-\U0001F16B\U0001F170-\U0001F1AC\U0001F1E6-\U0001F202\U0001F210-\U0001F23B\U0001F240-\U0001F248\U0001F250\U0001F251\U0001F260-\U0001F265\U0001F300-\U0001F3FA\U0001F400-\U0001F6D4\U0001F6E0-\U0001F6EC\U0001F6F0-\U0001F6F9\U0001F700-\U0001F773\U0001F780-\U0001F7D8\U0001F800-\U0001F80B\U0001F810-\U0001F847\U0001F850-\U0001F859\U0001F860-\U0001F887\U0001F890-\U0001F8AD\U0001F900-\U0001F90B\U0001F910-\U0001F93E\U0001F940-\U0001F970\U0001F973-\U0001F976\U0001F97A\U0001F97C-\U0001F9A2\U0001F9B0-\U0001F9B9\U0001F9C0-\U0001F9C2\U0001F9D0-\U0001F9FF\U0001FA60-\U0001FA6D]�suffix_search�2�…$|……$|,$|:$|;$|\!$|\?$|¿$|؟$|¡$|\($|\)$|\[$|\]$|\{$|\}$|<$|>$|_$|#$|\*$|&$|。$|？$|！$|，$|、$|；$|：$|～$|·$|।$|،$|۔$|؛$|٪$|\.\.+$|…$|\'$|"$|”$|“$|`$|‘$|´$|’$|‚$|,$|„$|»$|«$|「$|」$|『$|』$|（$|）$|〔$|〕$|【$|】$|《$|》$|〈$|〉$|〈$|〉$|⟦$|⟧$|[\u00A6\u00A9\u00AE\u00B0\u0482\u058D\u058E\u060E\u060F\u06DE\u06E9\u06FD\u06FE\u07F6\u09FA\u0B70\u0BF3-\u0BF8\u0BFA\u0C7F\u0D4F\u0D79\u0F01-\u0F03\u0F13\u0F15-\u0F17\u0F1A-\u0F1F\u0F34\u0F36\u0F38\u0FBE-\u0FC5\u0FC7-\u0FCC\u0FCE\u0FCF\u0FD5-\u0FD8\u109E\u109F\u1390-\u1399\u1940\u19DE-\u19FF\u1B61-\u1B6A\u1B74-\u1B7C\u2100\u2101\u2103-\u2106\u2108\u2109\u2114\u2116\u2117\u211E-\u2123\u2125\u2127\u2129\u212E\u213A\u213B\u214A\u214C\u214D\u214F\u218A\u218B\u2195-\u2199\u219C-\u219F\u21A1\u21A2\u21A4\u21A5\u21A7-\u21AD\u21AF-\u21CD\u21D0\u21D1\u21D3\u21D5-\u21F3\u2300-\u2307\u230C-\u231F\u2322-\u2328\u232B-\u237B\u237D-\u239A\u23B4-\u23DB\u23E2-\u2426\u2440-\u244A\u249C-\u24E9\u2500-\u25B6\u25B8-\u25C0\u25C2-\u25F7\u2600-\u266E\u2670-\u2767\u2794-\u27BF\u2800-\u28FF\u2B00-\u2B2F\u2B45\u2B46\u2B4D-\u2B73\u2B76-\u2B95\u2B98-\u2BC8\u2BCA-\u2BFE\u2CE5-\u2CEA\u2E80-\u2E99\u2E9B-\u2EF3\u2F00-\u2FD5\u2FF0-\u2FFB\u3004\u3012\u3013\u3020\u3036\u3037\u303E\u303F\u3190\u3191\u3196-\u319F\u31C0-\u31E3\u3200-\u321E\u322A-\u3247\u3250\u3260-\u327F\u328A-\u32B0\u32C0-\u32FE\u3300-\u33FF\u4DC0-\u4DFF\uA490-\uA4C6\uA828-\uA82B\uA836\uA837\uA839\uAA77-\uAA79\uFDFD\uFFE4\uFFE8\uFFED\uFFEE\uFFFC\uFFFD\U00010137-\U0001013F\U00010179-\U00010189\U0001018C-\U0001018E\U00010190-\U0001019B\U000101A0\U000101D0-\U000101FC\U00010877\U00010878\U00010AC8\U0001173F\U00016B3C-\U00016B3F\U00016B45\U0001BC9C\U0001D000-\U0001D0F5\U0001D100-\U0001D126\U0001D129-\U0001D164\U0001D16A-\U0001D16C\U0001D183\U0001D184\U0001D18C-\U0001D1A9\U0001D1AE-\U0001D1E8\U0001D200-\U0001D241\U0001D245\U0001D300-\U0001D356\U0001D800-\U0001D9FF\U0001DA37-\U0001DA3A\U0001DA6D-\U0001DA74\U0001DA76-\U0001DA83\U0001DA85\U0001DA86\U0001ECAC\U0001F000-\U0001F02B\U0001F030-\U0001F093\U0001F0A0-\U0001F0AE\U0001F0B1-\U0001F0BF\U0001F0C1-\U0001F0CF\U0001F0D1-\U0001F0F5\U0001F110-\U0001F16B\U0001F170-\U0001F1AC\U0001F1E6-\U0001F202\U0001F210-\U0001F23B\U0001F240-\U0001F248\U0001F250\U0001F251\U0001F260-\U0001F265\U0001F300-\U0001F3FA\U0001F400-\U0001F6D4\U0001F6E0-\U0001F6EC\U0001F6F0-\U0001F6F9\U0001F700-\U0001F773\U0001F780-\U0001F7D8\U0001F800-\U0001F80B\U0001F810-\U0001F847\U0001F850-\U0001F859\U0001F860-\U0001F887\U0001F890-\U0001F8AD\U0001F900-\U0001F90B\U0001F910-\U0001F93E\U0001F940-\U0001F970\U0001F973-\U0001F976\U0001F97A\U0001F97C-\U0001F9A2\U0001F9B0-\U0001F9B9\U0001F9C0-\U0001F9C2\U0001F9D0-\U0001F9FF\U0001FA60-\U0001FA6D]$|(?<=[0-9])\+$|([0-9])+\'$|([A-Za-z])?\'$|^([0-9]){1,2}\.$| ([0-9]){1,2}\.$|([0-9]){1}\) $|^([0-9]){1}\)$$|(?<=°[FfCcKk])\.$|([0-9])+\&$|(?<=[0-9])(?:\$|£|€|¥|฿|US\$|C\$|A\$|₽|﷼|₴|₠|₡|₢|₣|₤|₥|₦|₧|₨|₩|₪|₫|€|₭|₮|₯|₰|₱|₲|₳|₴|₵|₶|₷|₸|₹|₺|₻|₼|₽|₾|₿)$|(?<=[0-9])(?:km|km²|km³|m|m²|m³|dm|dm²|dm³|cm|cm²|cm³|mm|mm²|mm³|ha|µm|nm|yd|in|ft|kg|g|mg|µg|t|lb|oz|m/s|km/h|kmh|mph|hPa|Pa|mbar|mb|MB|kb|KB|gb|GB|tb|TB|T|G|M|K|км|км²|км³|м|м²|м³|дм|дм²|дм³|см|см²|см³|мм|мм²|мм³|нм|кг|г|мг|м/с|км/ч|кПа|Па|мбар|Кб|КБ|кб|Мб|МБ|мб|Гб|ГБ|гб|Тб|ТБ|тб)$|(?<=[0-9a-z\uFF41-\uFF5A\u00DF-\u00F6\u00F8-\u00FF\u0101\u0103\u0105\u0107\u0109\u010B\u010D\u010F\u0111\u0113\u0115\u0117\u0119\u011B\u011D\u011F\u0121\u0123\u0125\u0127\u0129\u012B\u012D\u012F\u0131\u0133\u0135\u0137\u0138\u013A\u013C\u013E\u0140\u0142\u0144\u0146\u0148\u0149\u014B\u014D\u014F\u0151\u0153\u0155\u0157\u0159\u015B\u015D\u015F\u0161\u0163\u0165\u0167\u0169\u016B\u016D\u016F\u0171\u0173\u0175\u0177\u017A\u017C\u017E\u017F\u0180\u0183\u0185\u0188\u018C\u018D\u0192\u0195\u0199-\u019B\u019E\u01A1\u01A3\u01A5\u01A8\u01AA\u01AB\u01AD\u01B0\u01B4\u01B6\u01B9\u01BA\u01BD-\u01BF\u01C6\u01C9\u01CC\u01CE\u01D0\u01D2\u01D4\u01D6\u01D8\u01DA\u01DC\u01DD\u01DF\u01E1\u01E3\u01E5\u01E7\u01E9\u01EB\u01ED\u01EF\u01F0\u01F3\u01F5\u01F9\u01FB\u01FD\u01FF\u0201\u0203\u0205\u0207\u0209\u020B\u020D\u020F\u0211\u0213\u0215\u0217\u0219\u021B\u021D\u021F\u0221\u0223\u0225\u0227\u0229\u022B\u022D\u022F\u0231\u0233-\u0239\u023C\u023F\u0240\u0242\u0247\u0249\u024B\u024D\u024F\u2C61\u2C65\u2C66\u2C68\u2C6A\u2C6C\u2C71\u2C73\u2C74\u2C76-\u2C7B\uA723\uA725\uA727\uA729\uA72B\uA72D\uA72F-\uA731\uA733\uA735\uA737\uA739\uA73B\uA73D\uA73F\uA741\uA743\uA745\uA747\uA749\uA74B\uA74D\uA74F\uA751\uA753\uA755\uA757\uA759\uA75B\uA75D\uA75F\uA761\uA763\uA765\uA767\uA769\uA76B\uA76D\uA76F\uA771-\uA778\uA77A\uA77C\uA77F\uA781\uA783\uA785\uA787\uA78C\uA78E\uA791\uA793-\uA795\uA797\uA799\uA79B\uA79D\uA79F\uA7A1\uA7A3\uA7A5\uA7A7\uA7A9\uA7AF\uA7B5\uA7B7\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E01\u1E03\u1E05\u1E07\u1E09\u1E0B\u1E0D\u1E0F\u1E11\u1E13\u1E15\u1E17\u1E19\u1E1B\u1E1D\u1E1F\u1E21\u1E23\u1E25\u1E27\u1E29\u1E2B\u1E2D\u1E2F\u1E31\u1E33\u1E35\u1E37\u1E39\u1E3B\u1E3D\u1E3F\u1E41\u1E43\u1E45\u1E47\u1E49\u1E4B\u1E4D\u1E4F\u1E51\u1E53\u1E55\u1E57\u1E59\u1E5B\u1E5D\u1E5F\u1E61\u1E63\u1E65\u1E67\u1E69\u1E6B\u1E6D\u1E6F\u1E71\u1E73\u1E75\u1E77\u1E79\u1E7B\u1E7D\u1E7F\u1E81\u1E83\u1E85\u1E87\u1E89\u1E8B\u1E8D\u1E8F\u1E91\u1E93\u1E95-\u1E9D\u1E9F\u1EA1\u1EA3\u1EA5\u1EA7\u1EA9\u1EAB\u1EAD\u1EAF\u1EB1\u1EB3\u1EB5\u1EB7\u1EB9\u1EBB\u1EBD\u1EBF\u1EC1\u1EC3\u1EC5\u1EC7\u1EC9\u1ECB\u1ECD\u1ECF\u1ED1\u1ED3\u1ED5\u1ED7\u1ED9\u1EDB\u1EDD\u1EDF\u1EE1\u1EE3\u1EE5\u1EE7\u1EE9\u1EEB\u1EED\u1EEF\u1EF1\u1EF3\u1EF5\u1EF7\u1EF9\u1EFB\u1EFD\u1EFFёа-яәөүҗңһα-ωάέίόώήύа-щюяіїєґѓѕјљњќѐѝ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F²\-\+(?:\'"”“`‘´’‚,„»«「」『』（）〔〕【】《》〈〉〈〉⟦⟧)])\.$|(?<=[A-Z\uFF21-\uFF3A\u00C0-\u00D6\u00D8-\u00DE\u0100\u0102\u0104\u0106\u0108\u010A\u010C\u010E\u0110\u0112\u0114\u0116\u0118\u011A\u011C\u011E\u0120\u0122\u0124\u0126\u0128\u012A\u012C\u012E\u0130\u0132\u0134\u0136\u0139\u013B\u013D\u013F\u0141\u0143\u0145\u0147\u014A\u014C\u014E\u0150\u0152\u0154\u0156\u0158\u015A\u015C\u015E\u0160\u0162\u0164\u0166\u0168\u016A\u016C\u016E\u0170\u0172\u0174\u0176\u0178\u0179\u017B\u017D\u0181\u0182\u0184\u0186\u0187\u0189-\u018B\u018E-\u0191\u0193\u0194\u0196-\u0198\u019C\u019D\u019F\u01A0\u01A2\u01A4\u01A6\u01A7\u01A9\u01AC\u01AE\u01AF\u01B1-\u01B3\u01B5\u01B7\u01B8\u01BC\u01C4\u01C7\u01CA\u01CD\u01CF\u01D1\u01D3\u01D5\u01D7\u01D9\u01DB\u01DE\u01E0\u01E2\u01E4\u01E6\u01E8\u01EA\u01EC\u01EE\u01F1\u01F4\u01F6-\u01F8\u01FA\u01FC\u01FE\u0200\u0202\u0204\u0206\u0208\u020A\u020C\u020E\u0210\u0212\u0214\u0216\u0218\u021A\u021C\u021E\u0220\u0222\u0224\u0226\u0228\u022A\u022C\u022E\u0230\u0232\u023A\u023B\u023D\u023E\u0241\u0243-\u0246\u0248\u024A\u024C\u024E\u2C60\u2C62-\u2C64\u2C67\u2C69\u2C6B\u2C6D-\u2C70\u2C72\u2C75\u2C7E\u2C7F\uA722\uA724\uA726\uA728\uA72A\uA72C\uA72E\uA732\uA734\uA736\uA738\uA73A\uA73C\uA73E\uA740\uA742\uA744\uA746\uA748\uA74A\uA74C\uA74E\uA750\uA752\uA754\uA756\uA758\uA75A\uA75C\uA75E\uA760\uA762\uA764\uA766\uA768\uA76A\uA76C\uA76E\uA779\uA77B\uA77D\uA77E\uA780\uA782\uA784\uA786\uA78B\uA78D\uA790\uA792\uA796\uA798\uA79A\uA79C\uA79E\uA7A0\uA7A2\uA7A4\uA7A6\uA7A8\uA7AA-\uA7AE\uA7B0-\uA7B4\uA7B6\uA7B8\u1E00\u1E02\u1E04\u1E06\u1E08\u1E0A\u1E0C\u1E0E\u1E10\u1E12\u1E14\u1E16\u1E18\u1E1A\u1E1C\u1E1E\u1E20\u1E22\u1E24\u1E26\u1E28\u1E2A\u1E2C\u1E2E\u1E30\u1E32\u1E34\u1E36\u1E38\u1E3A\u1E3C\u1E3E\u1E40\u1E42\u1E44\u1E46\u1E48\u1E4A\u1E4C\u1E4E\u1E50\u1E52\u1E54\u1E56\u1E58\u1E5A\u1E5C\u1E5E\u1E60\u1E62\u1E64\u1E66\u1E68\u1E6A\u1E6C\u1E6E\u1E70\u1E72\u1E74\u1E76\u1E78\u1E7A\u1E7C\u1E7E\u1E80\u1E82\u1E84\u1E86\u1E88\u1E8A\u1E8C\u1E8E\u1E90\u1E92\u1E94\u1E9E\u1EA0\u1EA2\u1EA4\u1EA6\u1EA8\u1EAA\u1EAC\u1EAE\u1EB0\u1EB2\u1EB4\u1EB6\u1EB8\u1EBA\u1EBC\u1EBE\u1EC0\u1EC2\u1EC4\u1EC6\u1EC8\u1ECA\u1ECC\u1ECE\u1ED0\u1ED2\u1ED4\u1ED6\u1ED8\u1EDA\u1EDC\u1EDE\u1EE0\u1EE2\u1EE4\u1EE6\u1EE8\u1EEA\u1EEC\u1EEE\u1EF0\u1EF2\u1EF4\u1EF6\u1EF8\u1EFA\u1EFC\u1EFEЁА-ЯӘӨҮҖҢҺΑ-ΩΆΈΊΌΏΉΎА-ЩЮЯІЇЄҐЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F][A-Z\uFF21-\uFF3A\u00C0-\u00D6\u00D8-\u00DE\u0100\u0102\u0104\u0106\u0108\u010A\u010C\u010E\u0110\u0112\u0114\u0116\u0118\u011A\u011C\u011E\u0120\u0122\u0124\u0126\u0128\u012A\u012C\u012E\u0130\u0132\u0134\u0136\u0139\u013B\u013D\u013F\u0141\u0143\u0145\u0147\u014A\u014C\u014E\u0150\u0152\u0154\u0156\u0158\u015A\u015C\u015E\u0160\u0162\u0164\u0166\u0168\u016A\u016C\u016E\u0170\u0172\u0174\u0176\u0178\u0179\u017B\u017D\u0181\u0182\u0184\u0186\u0187\u0189-\u018B\u018E-\u0191\u0193\u0194\u0196-\u0198\u019C\u019D\u019F\u01A0\u01A2\u01A4\u01A6\u01A7\u01A9\u01AC\u01AE\u01AF\u01B1-\u01B3\u01B5\u01B7\u01B8\u01BC\u01C4\u01C7\u01CA\u01CD\u01CF\u01D1\u01D3\u01D5\u01D7\u01D9\u01DB\u01DE\u01E0\u01E2\u01E4\u01E6\u01E8\u01EA\u01EC\u01EE\u01F1\u01F4\u01F6-\u01F8\u01FA\u01FC\u01FE\u0200\u0202\u0204\u0206\u0208\u020A\u020C\u020E\u0210\u0212\u0214\u0216\u0218\u021A\u021C\u021E\u0220\u0222\u0224\u0226\u0228\u022A\u022C\u022E\u0230\u0232\u023A\u023B\u023D\u023E\u0241\u0243-\u0246\u0248\u024A\u024C\u024E\u2C60\u2C62-\u2C64\u2C67\u2C69\u2C6B\u2C6D-\u2C70\u2C72\u2C75\u2C7E\u2C7F\uA722\uA724\uA726\uA728\uA72A\uA72C\uA72E\uA732\uA734\uA736\uA738\uA73A\uA73C\uA73E\uA740\uA742\uA744\uA746\uA748\uA74A\uA74C\uA74E\uA750\uA752\uA754\uA756\uA758\uA75A\uA75C\uA75E\uA760\uA762\uA764\uA766\uA768\uA76A\uA76C\uA76E\uA779\uA77B\uA77D\uA77E\uA780\uA782\uA784\uA786\uA78B\uA78D\uA790\uA792\uA796\uA798\uA79A\uA79C\uA79E\uA7A0\uA7A2\uA7A4\uA7A6\uA7A8\uA7AA-\uA7AE\uA7B0-\uA7B4\uA7B6\uA7B8\u1E00\u1E02\u1E04\u1E06\u1E08\u1E0A\u1E0C\u1E0E\u1E10\u1E12\u1E14\u1E16\u1E18\u1E1A\u1E1C\u1E1E\u1E20\u1E22\u1E24\u1E26\u1E28\u1E2A\u1E2C\u1E2E\u1E30\u1E32\u1E34\u1E36\u1E38\u1E3A\u1E3C\u1E3E\u1E40\u1E42\u1E44\u1E46\u1E48\u1E4A\u1E4C\u1E4E\u1E50\u1E52\u1E54\u1E56\u1E58\u1E5A\u1E5C\u1E5E\u1E60\u1E62\u1E64\u1E66\u1E68\u1E6A\u1E6C\u1E6E\u1E70\u1E72\u1E74\u1E76\u1E78\u1E7A\u1E7C\u1E7E\u1E80\u1E82\u1E84\u1E86\u1E88\u1E8A\u1E8C\u1E8E\u1E90\u1E92\u1E94\u1E9E\u1EA0\u1EA2\u1EA4\u1EA6\u1EA8\u1EAA\u1EAC\u1EAE\u1EB0\u1EB2\u1EB4\u1EB6\u1EB8\u1EBA\u1EBC\u1EBE\u1EC0\u1EC2\u1EC4\u1EC6\u1EC8\u1ECA\u1ECC\u1ECE\u1ED0\u1ED2\u1ED4\u1ED6\u1ED8\u1EDA\u1EDC\u1EDE\u1EE0\u1EE2\u1EE4\u1EE6\u1EE8\u1EEA\u1EEC\u1EEE\u1EF0\u1EF2\u1EF4\u1EF6\u1EF8\u1EFA\u1EFC\u1EFEЁА-ЯӘӨҮҖҢҺΑ-ΩΆΈΊΌΏΉΎА-ЩЮЯІЇЄҐЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])\.$|(?<=[Α-Ωα-ωίϊΐόάέύϋΰήώ])\-$|(?<=[Α-Ωα-ωίϊΐόάέύϋΰήώ])\.$|^[Α-Ω]{1}\.$|\ [Α-Ω]{1}\.$|[ΈΆΊΑΌ-Ωα-ωίϊΐόάέύϋΰήώ]+([\-]([ΈΆΊΑΌ-Ωα-ωίϊΐόάέύϋΰήώ]+))+$|([0-9]+)mg$|([0-9]+)\.([0-9]+)m$�infix_finditer�?d\.\.+|…|[\u00A6\u00A9\u00AE\u00B0\u0482\u058D\u058E\u060E\u060F\u06DE\u06E9\u06FD\u06FE\u07F6\u09FA\u0B70\u0BF3-\u0BF8\u0BFA\u0C7F\u0D4F\u0D79\u0F01-\u0F03\u0F13\u0F15-\u0F17\u0F1A-\u0F1F\u0F34\u0F36\u0F38\u0FBE-\u0FC5\u0FC7-\u0FCC\u0FCE\u0FCF\u0FD5-\u0FD8\u109E\u109F\u1390-\u1399\u1940\u19DE-\u19FF\u1B61-\u1B6A\u1B74-\u1B7C\u2100\u2101\u2103-\u2106\u2108\u2109\u2114\u2116\u2117\u211E-\u2123\u2125\u2127\u2129\u212E\u213A\u213B\u214A\u214C\u214D\u214F\u218A\u218B\u2195-\u2199\u219C-\u219F\u21A1\u21A2\u21A4\u21A5\u21A7-\u21AD\u21AF-\u21CD\u21D0\u21D1\u21D3\u21D5-\u21F3\u2300-\u2307\u230C-\u231F\u2322-\u2328\u232B-\u237B\u237D-\u239A\u23B4-\u23DB\u23E2-\u2426\u2440-\u244A\u249C-\u24E9\u2500-\u25B6\u25B8-\u25C0\u25C2-\u25F7\u2600-\u266E\u2670-\u2767\u2794-\u27BF\u2800-\u28FF\u2B00-\u2B2F\u2B45\u2B46\u2B4D-\u2B73\u2B76-\u2B95\u2B98-\u2BC8\u2BCA-\u2BFE\u2CE5-\u2CEA\u2E80-\u2E99\u2E9B-\u2EF3\u2F00-\u2FD5\u2FF0-\u2FFB\u3004\u3012\u3013\u3020\u3036\u3037\u303E\u303F\u3190\u3191\u3196-\u319F\u31C0-\u31E3\u3200-\u321E\u322A-\u3247\u3250\u3260-\u327F\u328A-\u32B0\u32C0-\u32FE\u3300-\u33FF\u4DC0-\u4DFF\uA490-\uA4C6\uA828-\uA82B\uA836\uA837\uA839\uAA77-\uAA79\uFDFD\uFFE4\uFFE8\uFFED\uFFEE\uFFFC\uFFFD\U00010137-\U0001013F\U00010179-\U00010189\U0001018C-\U0001018E\U00010190-\U0001019B\U000101A0\U000101D0-\U000101FC\U00010877\U00010878\U00010AC8\U0001173F\U00016B3C-\U00016B3F\U00016B45\U0001BC9C\U0001D000-\U0001D0F5\U0001D100-\U0001D126\U0001D129-\U0001D164\U0001D16A-\U0001D16C\U0001D183\U0001D184\U0001D18C-\U0001D1A9\U0001D1AE-\U0001D1E8\U0001D200-\U0001D241\U0001D245\U0001D300-\U0001D356\U0001D800-\U0001D9FF\U0001DA37-\U0001DA3A\U0001DA6D-\U0001DA74\U0001DA76-\U0001DA83\U0001DA85\U0001DA86\U0001ECAC\U0001F000-\U0001F02B\U0001F030-\U0001F093\U0001F0A0-\U0001F0AE\U0001F0B1-\U0001F0BF\U0001F0C1-\U0001F0CF\U0001F0D1-\U0001F0F5\U0001F110-\U0001F16B\U0001F170-\U0001F1AC\U0001F1E6-\U0001F202\U0001F210-\U0001F23B\U0001F240-\U0001F248\U0001F250\U0001F251\U0001F260-\U0001F265\U0001F300-\U0001F3FA\U0001F400-\U0001F6D4\U0001F6E0-\U0001F6EC\U0001F6F0-\U0001F6F9\U0001F700-\U0001F773\U0001F780-\U0001F7D8\U0001F800-\U0001F80B\U0001F810-\U0001F847\U0001F850-\U0001F859\U0001F860-\U0001F887\U0001F890-\U0001F8AD\U0001F900-\U0001F90B\U0001F910-\U0001F93E\U0001F940-\U0001F970\U0001F973-\U0001F976\U0001F97A\U0001F97C-\U0001F9A2\U0001F9B0-\U0001F9B9\U0001F9C0-\U0001F9C2\U0001F9D0-\U0001F9FF\U0001FA60-\U0001FA6D]|(?<=[0-9])[+\/\-\*^](?=[0-9])|([a-zA-Z]+)\/([a-zA-Z]+)\/([a-zA-Z]+)|([0-9])+(\.([0-9]+))*([\-]([0-9])+)+|([0-9])+[,]([0-9])+[\-]([0-9])+[,]([0-9])+|([0-9])+[ης]+([\-]([0-9])+)+|([0-9]){1,4}[\/]([0-9]){1,2}([\/]([0-9]){0,4}){0,1}|[A-Za-z]+\@[A-Za-z]+(\-[A-Za-z]+)*\.[A-Za-z]+|([a-zA-Z]+)(\-([a-zA-Z]+))+|(?<=[a-z\uFF41-\uFF5A\u00DF-\u00F6\u00F8-\u00FF\u0101\u0103\u0105\u0107\u0109\u010B\u010D\u010F\u0111\u0113\u0115\u0117\u0119\u011B\u011D\u011F\u0121\u0123\u0125\u0127\u0129\u012B\u012D\u012F\u0131\u0133\u0135\u0137\u0138\u013A\u013C\u013E\u0140\u0142\u0144\u0146\u0148\u0149\u014B\u014D\u014F\u0151\u0153\u0155\u0157\u0159\u015B\u015D\u015F\u0161\u0163\u0165\u0167\u0169\u016B\u016D\u016F\u0171\u0173\u0175\u0177\u017A\u017C\u017E\u017F\u0180\u0183\u0185\u0188\u018C\u018D\u0192\u0195\u0199-\u019B\u019E\u01A1\u01A3\u01A5\u01A8\u01AA\u01AB\u01AD\u01B0\u01B4\u01B6\u01B9\u01BA\u01BD-\u01BF\u01C6\u01C9\u01CC\u01CE\u01D0\u01D2\u01D4\u01D6\u01D8\u01DA\u01DC\u01DD\u01DF\u01E1\u01E3\u01E5\u01E7\u01E9\u01EB\u01ED\u01EF\u01F0\u01F3\u01F5\u01F9\u01FB\u01FD\u01FF\u0201\u0203\u0205\u0207\u0209\u020B\u020D\u020F\u0211\u0213\u0215\u0217\u0219\u021B\u021D\u021F\u0221\u0223\u0225\u0227\u0229\u022B\u022D\u022F\u0231\u0233-\u0239\u023C\u023F\u0240\u0242\u0247\u0249\u024B\u024D\u024F\u2C61\u2C65\u2C66\u2C68\u2C6A\u2C6C\u2C71\u2C73\u2C74\u2C76-\u2C7B\uA723\uA725\uA727\uA729\uA72B\uA72D\uA72F-\uA731\uA733\uA735\uA737\uA739\uA73B\uA73D\uA73F\uA741\uA743\uA745\uA747\uA749\uA74B\uA74D\uA74F\uA751\uA753\uA755\uA757\uA759\uA75B\uA75D\uA75F\uA761\uA763\uA765\uA767\uA769\uA76B\uA76D\uA76F\uA771-\uA778\uA77A\uA77C\uA77F\uA781\uA783\uA785\uA787\uA78C\uA78E\uA791\uA793-\uA795\uA797\uA799\uA79B\uA79D\uA79F\uA7A1\uA7A3\uA7A5\uA7A7\uA7A9\uA7AF\uA7B5\uA7B7\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E01\u1E03\u1E05\u1E07\u1E09\u1E0B\u1E0D\u1E0F\u1E11\u1E13\u1E15\u1E17\u1E19\u1E1B\u1E1D\u1E1F\u1E21\u1E23\u1E25\u1E27\u1E29\u1E2B\u1E2D\u1E2F\u1E31\u1E33\u1E35\u1E37\u1E39\u1E3B\u1E3D\u1E3F\u1E41\u1E43\u1E45\u1E47\u1E49\u1E4B\u1E4D\u1E4F\u1E51\u1E53\u1E55\u1E57\u1E59\u1E5B\u1E5D\u1E5F\u1E61\u1E63\u1E65\u1E67\u1E69\u1E6B\u1E6D\u1E6F\u1E71\u1E73\u1E75\u1E77\u1E79\u1E7B\u1E7D\u1E7F\u1E81\u1E83\u1E85\u1E87\u1E89\u1E8B\u1E8D\u1E8F\u1E91\u1E93\u1E95-\u1E9D\u1E9F\u1EA1\u1EA3\u1EA5\u1EA7\u1EA9\u1EAB\u1EAD\u1EAF\u1EB1\u1EB3\u1EB5\u1EB7\u1EB9\u1EBB\u1EBD\u1EBF\u1EC1\u1EC3\u1EC5\u1EC7\u1EC9\u1ECB\u1ECD\u1ECF\u1ED1\u1ED3\u1ED5\u1ED7\u1ED9\u1EDB\u1EDD\u1EDF\u1EE1\u1EE3\u1EE5\u1EE7\u1EE9\u1EEB\u1EED\u1EEF\u1EF1\u1EF3\u1EF5\u1EF7\u1EF9\u1EFB\u1EFD\u1EFFёа-яәөүҗңһα-ωάέίόώήύа-щюяіїєґѓѕјљњќѐѝ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])\.(?=[A-Z\uFF21-\uFF3A\u00C0-\u00D6\u00D8-\u00DE\u0100\u0102\u0104\u0106\u0108\u010A\u010C\u010E\u0110\u0112\u0114\u0116\u0118\u011A\u011C\u011E\u0120\u0122\u0124\u0126\u0128\u012A\u012C\u012E\u0130\u0132\u0134\u0136\u0139\u013B\u013D\u013F\u0141\u0143\u0145\u0147\u014A\u014C\u014E\u0150\u0152\u0154\u0156\u0158\u015A\u015C\u015E\u0160\u0162\u0164\u0166\u0168\u016A\u016C\u016E\u0170\u0172\u0174\u0176\u0178\u0179\u017B\u017D\u0181\u0182\u0184\u0186\u0187\u0189-\u018B\u018E-\u0191\u0193\u0194\u0196-\u0198\u019C\u019D\u019F\u01A0\u01A2\u01A4\u01A6\u01A7\u01A9\u01AC\u01AE\u01AF\u01B1-\u01B3\u01B5\u01B7\u01B8\u01BC\u01C4\u01C7\u01CA\u01CD\u01CF\u01D1\u01D3\u01D5\u01D7\u01D9\u01DB\u01DE\u01E0\u01E2\u01E4\u01E6\u01E8\u01EA\u01EC\u01EE\u01F1\u01F4\u01F6-\u01F8\u01FA\u01FC\u01FE\u0200\u0202\u0204\u0206\u0208\u020A\u020C\u020E\u0210\u0212\u0214\u0216\u0218\u021A\u021C\u021E\u0220\u0222\u0224\u0226\u0228\u022A\u022C\u022E\u0230\u0232\u023A\u023B\u023D\u023E\u0241\u0243-\u0246\u0248\u024A\u024C\u024E\u2C60\u2C62-\u2C64\u2C67\u2C69\u2C6B\u2C6D-\u2C70\u2C72\u2C75\u2C7E\u2C7F\uA722\uA724\uA726\uA728\uA72A\uA72C\uA72E\uA732\uA734\uA736\uA738\uA73A\uA73C\uA73E\uA740\uA742\uA744\uA746\uA748\uA74A\uA74C\uA74E\uA750\uA752\uA754\uA756\uA758\uA75A\uA75C\uA75E\uA760\uA762\uA764\uA766\uA768\uA76A\uA76C\uA76E\uA779\uA77B\uA77D\uA77E\uA780\uA782\uA784\uA786\uA78B\uA78D\uA790\uA792\uA796\uA798\uA79A\uA79C\uA79E\uA7A0\uA7A2\uA7A4\uA7A6\uA7A8\uA7AA-\uA7AE\uA7B0-\uA7B4\uA7B6\uA7B8\u1E00\u1E02\u1E04\u1E06\u1E08\u1E0A\u1E0C\u1E0E\u1E10\u1E12\u1E14\u1E16\u1E18\u1E1A\u1E1C\u1E1E\u1E20\u1E22\u1E24\u1E26\u1E28\u1E2A\u1E2C\u1E2E\u1E30\u1E32\u1E34\u1E36\u1E38\u1E3A\u1E3C\u1E3E\u1E40\u1E42\u1E44\u1E46\u1E48\u1E4A\u1E4C\u1E4E\u1E50\u1E52\u1E54\u1E56\u1E58\u1E5A\u1E5C\u1E5E\u1E60\u1E62\u1E64\u1E66\u1E68\u1E6A\u1E6C\u1E6E\u1E70\u1E72\u1E74\u1E76\u1E78\u1E7A\u1E7C\u1E7E\u1E80\u1E82\u1E84\u1E86\u1E88\u1E8A\u1E8C\u1E8E\u1E90\u1E92\u1E94\u1E9E\u1EA0\u1EA2\u1EA4\u1EA6\u1EA8\u1EAA\u1EAC\u1EAE\u1EB0\u1EB2\u1EB4\u1EB6\u1EB8\u1EBA\u1EBC\u1EBE\u1EC0\u1EC2\u1EC4\u1EC6\u1EC8\u1ECA\u1ECC\u1ECE\u1ED0\u1ED2\u1ED4\u1ED6\u1ED8\u1EDA\u1EDC\u1EDE\u1EE0\u1EE2\u1EE4\u1EE6\u1EE8\u1EEA\u1EEC\u1EEE\u1EF0\u1EF2\u1EF4\u1EF6\u1EF8\u1EFA\u1EFC\u1EFEЁА-ЯӘӨҮҖҢҺΑ-ΩΆΈΊΌΏΉΎА-ЩЮЯІЇЄҐЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])|(?<=[A-Za-z\uFF21-\uFF3A\uFF41-\uFF5A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u017F\u0180-\u01BF\u01C4-\u024F\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFFёа-яЁА-ЯәөүҗңһӘӨҮҖҢҺα-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎа-щюяіїєґА-ЩЮЯІЇЄҐѓѕјљњќѐѝЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F]),(?=[A-Za-z\uFF21-\uFF3A\uFF41-\uFF5A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u017F\u0180-\u01BF\u01C4-\u024F\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFFёа-яЁА-ЯәөүҗңһӘӨҮҖҢҺα-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎа-щюяіїєґА-ЩЮЯІЇЄҐѓѕјљњќѐѝЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])|(?<=[A-Za-z\uFF21-\uFF3A\uFF41-\uFF5A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u017F\u0180-\u01BF\u01C4-\u024F\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFFёа-яЁА-ЯәөүҗңһӘӨҮҖҢҺα-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎа-щюяіїєґА-ЩЮЯІЇЄҐѓѕјљњќѐѝЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])(?:-|–|—|--|---|——|~)(?=[A-Za-z\uFF21-\uFF3A\uFF41-\uFF5A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u017F\u0180-\u01BF\u01C4-\u024F\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFFёа-яЁА-ЯәөүҗңһӘӨҮҖҢҺα-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎа-щюяіїєґА-ЩЮЯІЇЄҐѓѕјљњќѐѝЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])|(?<=[A-Za-z\uFF21-\uFF3A\uFF41-\uFF5A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u017F\u0180-\u01BF\u01C4-\u024F\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFFёа-яЁА-ЯәөүҗңһӘӨҮҖҢҺα-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎа-щюяіїєґА-ЩЮЯІЇЄҐѓѕјљњќѐѝЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])[:<>=/](?=[A-Za-z\uFF21-\uFF3A\uFF41-\uFF5A\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF\u0100-\u017F\u0180-\u01BF\u01C4-\u024F\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFFёа-яЁА-ЯәөүҗңһӘӨҮҖҢҺα-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎа-щюяіїєґА-ЩЮЯІЇЄҐѓѕјљњќѐѝЃЅЈЉЊЌЀЍ\u1200-\u137F\u0980-\u09FF\u0591-\u05F4\uFB1D-\uFB4F\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB\u0D80-\u0DFF\u0900-\u097F\u0C80-\u0CFF\u0B80-\u0BFF\u0C00-\u0C7F\uAC00-\uD7AF\u1100-\u11FF\u3040-\u309F\u30A0-\u30FFー\u4E00-\u62FF\u6300-\u77FF\u7800-\u8CFF\u8D00-\u9FFF\u3400-\u4DBF\U00020000-\U000215FF\U00021600-\U000230FF\U00023100-\U000245FF\U00024600-\U000260FF\U00026100-\U000275FF\U00027600-\U000290FF\U00029100-\U0002A6DF\U0002A700-\U0002B73F\U0002B740-\U0002B81F\U0002B820-\U0002CEAF\U0002CEB0-\U0002EBEF\u2E80-\u2EFF\u2F00-\u2FDF\u2FF0-\u2FFF\u3000-\u303F\u31C0-\u31EF\u3200-\u32FF\u3300-\u33FF\uF900-\uFAFF\uFE30-\uFE4F\U0001F200-\U0001F2FF\U0002F800-\U0002FA1F])�token_match��url_match�
2
+ ��A�
3
+ � ��A� �$ΗΠΑ��A�$ΗΠΑ�'��A�'�''��A�''�'ΣΟΥΝ��A�'ΣΟΥΝC�είμαι�'βγαινε��A�'βγαινεC�βγαίνω�'βλεπε��A�'βλεπεC�βλέπω�'γραφα��A�'γραφαC�γράφω�'γω��A�'γωC�εγώ�'δειξε��A�'δειξεC�δείχνω�'θελα��A�'θελαC�θέλω�'κανε��A�'κανεC�κάνω�'κει��A�'κειC�εκεί�'κλεβε��A�'κλεβεC�κλέβω�'λεγα��A�'λεγαC�λέγω�'λεγαν��A�'λεγανC�λέγω�'λεγε��A�'λεγεC�λέγω�'μας��A�'μαςC�εμάς�'μαστε��A�'μαστεC�είμαι�'μεις��A�'μειςC�εγώ�'μετ��A�'μετC�μετά�'μουν��A�'μουνC�είμαι�'μουνα��A�'μουναC�είμαι�'ναι��A�'ναιC�είμαι�'ξερες��A�'ξερεςC�ξέρω�'ξομολογήθηκε��A�'ξομολογήθηκεC�εξομολογούμαι�'παιρνα��A�'παιρναC�παίρνω�'πανε��A�'πανεC�λέγω�'πε��A�'πεC�λέγω�'πρεπε��A�'πρεπεC�πρέπει�'ρθε��A�'ρθεC�έρχομαι�'ρθεί��A�'ρθείC�έρχομαι�'ρθει��A�'ρθειC�έρχομαι�'ρθουν��A�'ρθουνC�έρχομαι�'ρθούνε��A�'ρθούνεC�έρχομαι�'ριξε��A�'ριξεC�ρίχνω�'ρχεται��A�'ρχεταιC�έρχομαι�'σένα��A�'σέναC�εγώ�'σβηνε��A�'σβηνεC�σβήνω�'σκασε��A�'σκασεC�σκάω�'σπαγαν��A�'σπαγανC�σπάω�'ταν��A�'τανC�είμαι�'τανε��A�'τανεC�είμαι�'φαγε��A�'φαγεC�τρώω�'φερε��A�'φερεC�φέρνω�'φύγε��A�'φύγεC�φεύγω�'χα��A�'χαC�έχω�'χαμε��A�'χαμεC�έχω�'χε��A�'χεC�έχω�'χει��A�'χειC�έχω�'χεις��A�'χειςC�έχω�'χουμε��A�'χουμεC�έχω�'χουν��A�'χουνC�έχω�(*_*)��A�(*_*)�(-8��A�(-8�(-:��A�(-:�(-;��A�(-;�(-_-)��A�(-_-)�(._.)��A�(._.)�(:��A�(:�(;��A�(;�(=��A�(=�(>_<)��A�(>_<)�(^_^)��A�(^_^)�(o:��A�(o:�(¬_¬)��A�(¬_¬)�(ಠ_ಠ)��A�(ಠ_ಠ)�(╯°□°）╯︵┻━┻��A�(╯°□°）╯︵┻━┻�)-:��A�)-:�):��A�):�-_-��A�-_-�-__-��A�-__-�._.��A�._.�0.0��A�0.0�0.o��A�0.o�0_0��A�0_0�0_o��A�0_o�10μ.μ.��A�10�A�μ.μ.C�μ.μ.�10μμ��A�10�A�μμC�μ.μ.�10π.μ.��A�10�A�π.μ.C�π.μ.�10πμ��A�10�A�πμC�π.μ.�11μ.μ.��A�11�A�μ.μ.C�μ.μ.�11μμ��A�11�A�μμC�μ.μ.�11π.μ.��A�11�A�π.μ.C�π.μ.�11πμ��A�11�A�πμC�π.μ.�12μ.μ.��A�12�A�μ.μ.C�μ.μ.�12μμ��A�12�A�μμC�μ.μ.�12π.μ.��A�12�A�π.μ.C�π.μ.�12πμ��A�12�A�πμC�π.μ.�1μ.μ.��A�1�A�μ.μ.C�μ.μ.�1μμ��A�1�A�μμC�μ.μ.�1π.μ.��A�1�A�π.μ.C�π.μ.�1πμ��A�1�A�πμC�π.μ.�2μ.μ.��A�2�A�μ.μ.C�μ.μ.�2μμ��A�2�A�μμC�μ.μ.�2π.μ.��A�2�A�π.μ.C�π.μ.�2πμ��A�2�A�πμC�π.μ.�3μ.μ.��A�3�A�μ.μ.C�μ.μ.�3μμ��A�3�A�μμC�μ.μ.�3π.μ.��A�3�A�π.μ.C�π.μ.�3πμ��A�3�A�πμC�π.μ.�4μ.μ.��A�4�A�μ.μ.C�μ.μ.�4μμ��A�4�A�μμC�μ.μ.�4π.μ.��A�4�A�π.μ.C�π.μ.�4πμ��A�4�A�πμC�π.μ.�5μ.μ.��A�5�A�μ.μ.C�μ.μ.�5μμ��A�5�A�μμC�μ.μ.�5π.μ.��A�5�A�π.μ.C�π.μ.�5πμ��A�5�A�πμC�π.μ.�6μ.μ.��A�6�A�μ.μ.C�μ.μ.�6μμ��A�6�A�μμC�μ.μ.�6π.μ.��A�6�A�π.μ.C�π.μ.�6πμ��A�6�A�πμC�π.μ.�7μ.μ.��A�7�A�μ.μ.C�μ.μ.�7μμ��A�7�A�μμC�μ.μ.�7π.μ.��A�7�A�π.μ.C�π.μ.�7πμ��A�7�A�πμC�π.μ.�8)��A�8)�8-)��A�8-)�8-D��A�8-D�8D��A�8D�8μ.μ.��A�8�A�μ.μ.C�μ.μ.�8μμ��A�8�A�μμC�μ.μ.�8π.μ.��A�8�A�π.μ.C�π.μ.�8πμ��A�8�A�πμC�π.μ.�9μ.μ.��A�9�A�μ.μ.C�μ.μ.�9μμ��A�9�A�μμC�μ.μ.�9π.μ.��A�9�A�π.μ.C�π.μ.�9πμ��A�9�A�πμC�π.μ.�:'(��A�:'(�:')��A�:')�:'-(��A�:'-(�:'-)��A�:'-)�:(��A�:(�:((��A�:((�:(((��A�:(((�:()��A�:()�:)��A�:)�:))��A�:))�:)))��A�:)))�:*��A�:*�:-(��A�:-(�:-((��A�:-((�:-(((��A�:-(((�:-)��A�:-)�:-))��A�:-))�:-)))��A�:-)))�:-*��A�:-*�:-/��A�:-/�:-0��A�:-0�:-3��A�:-3�:->��A�:->�:-D��A�:-D�:-O��A�:-O�:-P��A�:-P�:-X��A�:-X�:-]��A�:-]�:-o��A�:-o�:-p��A�:-p�:-x��A�:-x�:-|��A�:-|�:-}��A�:-}�:/��A�:/�:0��A�:0�:1��A�:1�:3��A�:3�:>��A�:>�:D��A�:D�:O��A�:O�:P��A�:P�:X��A�:X�:]��A�:]�:o��A�:o�:o)��A�:o)�:p��A�:p�:x��A�:x�:|��A�:|�:}��A�:}�:’(��A�:’(�:’)��A�:’)�:’-(��A�:’-(�:’-)��A�:’-)�;)��A�;)�;-)��A�;-)�;-D��A�;-D�;D��A�;D�;_;��A�;_;�<.<��A�<.<�</3��A�</3�<3��A�<3�<33��A�<33�<333��A�<333�<space>��A�<space>�=(��A�=(�=)��A�=)�=/��A�=/�=3��A�=3�=D��A�=D�=[��A�=[�=]��A�=]�=|��A�=|�>.<��A�>.<�>.>��A�>.>�>:(��A�>:(�>:o��A�>:o�><(((*>��A�><(((*>�@_@��A�@_@�C++��A�C++�O.O��A�O.O�O.o��A�O.o�O_O��A�O_O�O_o��A�O_o�V.V��A�V.V�V_V��A�V_V�XD��A�XD�XDD��A�XDD�[-:��A�[-:�[:��A�[:�[=��A�[=�\")��A�\")�\n��A�\n�\t��A�\t�]=��A�]=�^_^��A�^_^�^__^��A�^__^�^___^��A�^___^�a.��A�a.�b.��A�b.�c.��A�c.�d.��A�d.�e.��A�e.�f.��A�f.�g.��A�g.�h.��A�h.�i.��A�i.�j.��A�j.�k.��A�k.�l.��A�l.�m.��A�m.�n.��A�n.�o.��A�o.�o.0��A�o.0�o.O��A�o.O�o.o��A�o.o�o_0��A�o_0�o_O��A�o_O�o_o��A�o_o�p.��A�p.�q.��A�q.�r.��A�r.�s.��A�s.�t.��A�t.�u.��A�u.�v.��A�v.�v.v��A�v.v�v_v��A�v_v�w.��A�w.�x.��A�x.�xD��A�xD�xDD��A�xDD�y.��A�y.�z.��A�z.� ��A� C� �¯\(ツ)/¯��A�¯\(ツ)/¯�°C.��A�°�A�C�A�.�°F.��A�°�A�F�A�.�°K.��A�°�A�K�A�.�°c.��A�°�A�c�A�.�°f.��A�°�A�f�A�.�°k.��A�°�A�k�A�.�ä.��A�ä.�ö.��A�ö.�ü.��A�ü.�Α'��A�Α'�Α.Ε.��A�Α.Ε.�Α.Ε.Β.Ε.��A�Α.Ε.Β.Ε.�Α.Ε.Ι.��A�Α.Ε.Ι.�Α.Ε.Π.��A�Α.Ε.Π.�Α.Μ.Α.��A�Α.Μ.Α.�Α.Π.Θ.��A�Α.Π.Θ.�Α.Τ.��A�Α.Τ.�Α.Χ.��A�Α.Χ.�ΑΓΡ.��A�ΑΓΡ.C�Αγροτικός�ΑΝ.��A�ΑΝ.�ΑΠ'��A�ΑΠ'C�από�ΑΠ’��A�ΑΠ’C�από�Αγ.��A�Αγ.�Αγ. Γρ.��A�Αγ. Γρ.C�Αγία Γραφή�Αγόρασ'��A�Αγόρασ'C�αγοράζω�Αγόρασ’��A�Αγόρασ’C�αγοράζω�Αθ.��A�Αθ.C�Αθανάσιος�Αλ.��A�Αλ.�Αλεξ.��A�Αλεξ.C�Αλέξανδρος�Αλλ'��A�Αλλ'C�αλλά�Αλλ’��A�Αλλ’C�αλλά�Αμ'��A�Αμ'C�άμα�Αμ’��A�Αμ’C�άμα�Αν.��A�Αν.�Αντ'��A�Αντ'C�αντί�Αντ.��A�Αντ.�Αντ’��A�Αντ’C�αντί�Απ'��A�Απ'C�από�Απ.��A�Απ.�Απρ.��A�Απρ.C�Απρίλιος�Απ’��A�Απ’C�από�Αφ'��A�Αφ'C�από�Αφ’��A�Αφ’C�από�Αύγ.��A�Αύγ.C�Αύγουστος�Α’��A�Α’�Β'��A�Β'�Β)��A�Β)�Β.Ζ.��A�Β.Ζ.�Β.Ι.Ο.��A�Β.Ι.Ο.�Β.Κ.��A�Β.Κ.�Β.Μ.Α.��A�Β.Μ.Α.�Βάψ'��A�Βάψ'C�βάφω�Βάψ’��A�Βάψ’C�βάφω�Βασ.��A�Βασ.�Β’��A�Β’�Γ'��A�Γ'�Γ)��A�Γ)�Γ.Γ.��A�Γ.Γ.�Γ.Δ.��A�Γ.Δ.�ΓΙ'��A�ΓΙ'C�για�ΓΙ’��A�ΓΙ’C�για�Γι'��A�Γι'C�για�Γι’��A�Γι’C�για�Γκ.��A�Γκ.�Γ’��A�Γ’�Δ.Ε.Η.��A�Δ.Ε.Η.�Δ.Ε.Σ.Ε.��A�Δ.Ε.Σ.Ε.�Δ.Ν.��A�Δ.Ν.�Δ.Ο.Υ.��A�Δ.Ο.Υ.�Δ.Σ.��A�Δ.Σ.�Δ.Υ.��A�Δ.Υ.�ΔΙ.ΚΑ.Τ.Σ.Α.��A�ΔΙ.ΚΑ.Τ.Σ.Α.�Δέσ'��A�Δέσ'C�δένω�Δέσ’��A�Δέσ’C�δένω�Δεκ.��A�Δεκ.C�Δεκέμβριος�Δεσ'��A�Δεσ'C�δένω�Δεσ’��A�Δεσ’C�δένω�Δηλ.��A�Δηλ.�Δημ.��A�Δημ.C�Δήμος�Δι'��A�Δι'C�δια�Διον.��A�Διον.�Δι’��A�Δι’C�δια�Δωσ'��A�Δωσ'C�δίνω�Δωσ’��A�Δωσ’C�δίνω�Δύσκολ'��A�Δύσκολ'C�δύσκολος�Δύσκολ’��A�Δύσκολ’C�δύσκολος�Ε.Α.��A�Ε.Α.�Ε.Α.Κ.��A�Ε.Α.Κ.�Ε.Α.Π.��A�Ε.Α.Π.�Ε.Ε.��A�Ε.Ε.�Ε.Κ.��A�Ε.Κ.�Ε.ΚΕ.ΠΙΣ.��A�Ε.ΚΕ.ΠΙΣ.�Ε.Λ.Α.��A�Ε.Λ.Α.�Ε.Λ.Ι.Α.��A�Ε.Λ.Ι.Α.�Ε.Π.Σ.��A�Ε.Π.Σ.�Ε.Π.Τ.Α.��A�Ε.Π.Τ.Α.�Ε.Σ.Ε.Ε.Κ.��A�Ε.Σ.Ε.Ε.Κ.�Ε.Υ.Κ.��A�Ε.Υ.Κ.�ΕΕ.��A�ΕΕ.�ΕΚ.��A�ΕΚ.�ΕΛ.��A�ΕΛ.�ΕΛ.ΑΣ.��A�ΕΛ.ΑΣ.�Εθν.��A�Εθν.�Ελ.��A�Ελ.�Εμ.��A�Εμ.�Επ'��A�Επ'C�επί�Επ.��A�Επ.�Επ’��A�Επ’C�επί�Ευ.��A�Ευ.�Εφ'��A�Εφ'C�επί�Εφ’��A�Εφ’C�επί�Η'��A�Η'�Η.Π.Α.��A�Η.Π.Α.�Η’��A�Η’�Θ'��A�Θ'C�θα�ΘΕ.��A�ΘΕ.�Θεμ.��A�Θεμ.�Θεοδ.��A�Θεοδ.�Θρ.��A�Θρ.�Θ’��A�Θ’C�θα�Ι.Ε.Κ.��A�Ι.Ε.Κ.�Ι.Κ.Α.��A�Ι.Κ.Α.�Ι.Κ.Υ.��A�Ι.Κ.Υ.�Ι.Σ.Θ.��A�Ι.Σ.Θ.�Ι.Χ.��A�Ι.Χ.�ΙΖ'��A�ΙΖ'�ΙΖ’��A�ΙΖ’�ΙΧ.��A�ΙΧ.�Ιαν.��A�Ιαν.C�Ιανουάριος�Ιούλ.��A�Ιούλ.C�Ιούλιος�Ιούν.��A�Ιούν.C�Ιούνιος�Ιωαν.��A�Ιωαν.C�Ιωάννης�Κ.Α.Α.��A�Κ.Α.Α.�Κ.Α.Ε.��A�Κ.Α.Ε.�Κ.Β.Σ.��A�Κ.Β.Σ.�Κ.Δ.��A�Κ.Δ.�Κ.Ε.��A�Κ.Ε.�Κ.Ε.Κ.��A�Κ.Ε.Κ.�Κ.Ι.��A�Κ.Ι.�Κ.Ι.Θ.��A�Κ.Ι.Θ.�Κ.Κ.��A�Κ.Κ.�Κ.ΚΕΚ.��A�Κ.ΚΕΚ.�Κ.Ο.��A�Κ.Ο.�Κ.Π.Ρ.��A�Κ.Π.Ρ.�ΚΑΤ.��A�ΚΑΤ.�ΚΚ.��A�ΚΚ.�Κάν'��A�Κάν'C�κάνω�Κάν’��A�Κάν’C�κάνω�Καθ'��A�Καθ'C�κάθε�Καθ’��A�Καθ’C�κάθε�Καν.��A�Καν.�Καρ.��A�Καρ.�Κατ'��A�Κατ'C�κατά�Κατ.��A�Κατ.�Κατ’��A�Κατ’C�κατά�Κυρ.��A�Κυρ.�Κων.��A�Κων.�Λ.Α.��A�Λ.Α.�Λ.Χ.��A�Λ.Χ.�Λ.χ.��A�Λ.χ.�Λήξ'��A�Λήξ'C�λήγω�Λήξ’��A�Λήξ’C�λήγω�Λεωφ.��A�Λεωφ.�Λι.��A�Λι.�Μ'��A�Μ'C�με�Μ. Ασία��A�Μ. ΑσίαC�Μικρά Ασία�Μ.Δ.Ε.��A�Μ.Δ.Ε.�Μ.Ε.Ο.��A�Μ.Ε.Ο.�Μ.Ζ.��A�Μ.Ζ.�Μ.Μ.Ε.��A�Μ.Μ.Ε.�Μ.Ο.��A�Μ.Ο.�ΜΕΤ'��A�ΜΕΤ'C�μετά�ΜΕΤ’��A�ΜΕΤ’C�μετά�Μάρτ'��A�Μάρτ'C�Μάρτιος�Μάρτ.��A�Μάρτ.C�Μάρτιος�Μάρτ’��A�Μάρτ’C�Μάρτιος�Μέσ'��A�Μέσ'C�μέσα�Μέσ’��A�Μέσ’C�μέσα�Μεγ.��A�Μεγ.�Μετ'��A�Μετ'C�μετά�Μετ’��A�Μετ’C�μετά�Μιλτ.��A�Μιλτ.�Μιχ.��A�Μιχ.�Μ’��A�Μ’C�με�Ν'��A�Ν'C�να�Ν.Δ.��A�Ν.Δ.�Ν.Ε.Α.��A�Ν.Ε.Α.�Ν.Κ.��A�Ν.Κ.�Ν.Ο.��A�Ν.Ο.�Ν.Ο.Θ.��A�Ν.Ο.Θ.�Ν.Π.Δ.Δ.��A�Ν.Π.Δ.Δ.�Ν.Υ.��A�Ν.Υ.�ΝΔ.��A�ΝΔ.�Νικ.��A�Νικ.�Νοέμβρ.��A�Νοέμβρ.C�Νοέμβριος�Ντ'��A�Ντ'�Ντ.��A�Ντ.�Ντ’��A�Ντ’�Ν’��A�Ν’C�να�Ο'��A�Ο'�Ο.Α.��A�Ο.Α.�Ο.Α.Ε.Δ.��A�Ο.Α.Ε.Δ.�Ο.Δ.��A�Ο.Δ.�Ο.Ε.Ε.��A�Ο.Ε.Ε.�Ο.Ε.Ε.Κ.��A�Ο.Ε.Ε.Κ.�Ο.Η.Ε.��A�Ο.Η.Ε.�Ο.Κ.��A�Ο.Κ.�Οκτ.��A���κτ.C�Οκτώβριος�Ο’��A�Ο’�Π.Δ.��A�Π.Δ.�Π.Ε.Κ.Δ.Υ.��A�Π.Ε.Κ.Δ.Υ.�Π.Ε.Π.��A�Π.Ε.Π.�Π.Μ.Σ.��A�Π.Μ.Σ.�Π.Χ.��A�Π.Χ.�ΠΑΡ'��A�ΠΑΡ'C�παρά�ΠΑΡ’��A�ΠΑΡ’C�παρά�ΠΟΛ.��A�ΠΟΛ.�Πάρ'��A�Πάρ'C�παίρνω�Πάρ’��A�Πάρ’C�παίρνω�Παίξ'��A�Παίξ'C�παίζω�Παίξ’��A�Παίξ’C�παίζω�Παρ'��A�Παρ'C�παρά�Παρ.��A�Παρ.�Παρ’��A�Παρ’C�παρά�Πλ.��A�Πλ.�Πρ.��A�Πρ.�Σ'��A�Σ'C�σε�Σ.Δ.Ο.Ε.��A�Σ.Δ.Ο.Ε.�Σ.Ε.��A�Σ.Ε.�Σ.Ε.Κ.��A�Σ.Ε.Κ.�Σ.Π.Δ.Ω.Β.��A�Σ.Π.Δ.Ω.Β.�Σ.Τ.��A�Σ.Τ.�Σαβ.��A�Σαβ.�Σεπτ.��A�Σεπτ.C�Σεπτέμβριος�Στ'��A�Στ'C�στο�Στ.��A�Στ.�ΣτΕ.��A�ΣτΕ.�Στρ.��A�Στρ.�Στ’��A�Στ’C�στο�Σ’��A�Σ’C�σε�Τ'��A�Τ'C�το�Τ.Α.��A�Τ.Α.�Τ.Ε.Ε.��A�Τ.Ε.Ε.�Τ.Ε.Ι.��A�Τ.Ε.Ι.�ΤΡ.��A�ΤΡ.�Τζ.��A�Τζ.�Τηλ.��A�Τηλ.�Τ’��A�Τ’C�το�Υ.Γ.��A�Υ.Γ.�ΥΓ.��A�ΥΓ.�ΥΠ.Ε.Π.Θ.��A�ΥΠ.Ε.Π.Θ.�Υπ'��A�Υπ'C�υπό�Υπ’��A�Υπ’C�υπό�Φ.Α.Β.Ε.��A�Φ.Α.Β.Ε.�Φ.Κ.��A�Φ.Κ.�Φ.Π.Α.��A�Φ.Π.Α.�Φ.Σ.��A�Φ.Σ.�Φ.Χ.��A�Φ.Χ.�Φέρ'��A�Φέρ'C�φέρνω�Φέρ’��A�Φέρ’C�φέρνω�Φεβρ.��A�Φεβρ.C�Φεβρουάριος�Φερ'��A�Φερ'C�φέρνω�Φερ’��A�Φερ’C�φέρνω�Φιλ.��A�Φιλ.�Χ.Α.Α.��A�Χ.Α.Α.�Χ.Χ.��A�Χ.Χ.�ΧΡ.��A�ΧΡ.�Χαρ.��A�Χαρ.�Χιλ.��A�Χιλ.�Χρ.��A�Χρ.�άγ.��A�άγ.�άρθρ.��A�άρθρ.�άσ'��A�άσ'C�αφήνω�άσ’��A�άσ’C�αφήνω�έφθασ'��A�έφθασ'C�φθάνω�έφθασ’��A�έφθασ’C�φθάνω�αι.��A�αι.�αλλ'��A�αλλ'C�αλλά�αλλ’��A�αλλ’C�αλλά�αν.��A�αν.�απ.��A�απ.�αρ.��A�αρ.�αριθ.��A�αριθ.�αριθμ.��A�αριθμ.�αφ'��A�αφ'C�από�αφ’��A�αφ’C�από�β'��A�β'�βλ.��A�βλ.�β’��A�β’�γ.γ.��A�γ.γ.�γεν.��A�γεν.�γι'��A�γι'C�για�γι’��A�γι’C�για�γραμμ.��A�γραμμ.�δ.δ.��A�δ.δ.�δ.σ.��A�δ.σ.�δεσ'��A�δεσ'C�δένω�δεσ’��A�δεσ’C�δένω�δηλ.��A�δηλ.�δι'��A�δι'C�δια�δισ.��A�δισ.�δι’��A�δι’C�δια�δολ.��A�δολ.�δρχ.��A�δρχ.�δώσ'��A�δώσ'C�δίνω�δώσ’��A�δώσ’C�δίνω�εκ.��A�εκ.�εκατ.��A�εκατ.�ελ.��A�ελ.�εξ'��A�εξ'C�εκ�εξ’��A�εξ’C�εκ�επ'��A�επ'C�επί�επ’��A�επ’C�επί�εφ'��A�εφ'C�επί�εφ’��A�εφ’C�επί�θ'��A�θ'C�θα�θιν'��A�θιν'�θιν’��A�θιν’�θ’��A�θ’C�θα�κ'τσή��A�κ'τσήC�κουτσός�κ.��A�κ.�κ.ά.��A�κ.ά.�κ.α.��A�κ.α.�κ.κ.��A�κ.κ.�κ.λπ.��A�κ.λπ.�κ.ο.κ.��A�κ.ο.κ.�κ.τ.λ.��A�κ.τ.λ.�καθ'��A�καθ'C�κάθε�καθ’��A�καθ’C�κάθε�κατ'��A�κατ'C�κατά�κατ’��A�κατ’C�κατά�κι��A�κιC�και�κλπ.��A�κλπ.�κτλ.��A�κτλ.�κυβ.��A�κυβ.�κ’τσή��A�κ’τσήC�κουτσός�λ.χ.��A�λ.χ.�μ'��A�μ'C�με�μ.��A�μ.�μ.Χ.��A�μ.Χ.�μ.μ.��A�μ.μ.�μέσ'��A�μέσ'C�μέσα�μέσ’��A�μέσ’C�μέσα�μεσ'��A�μεσ'C�μέσα�μεσ’��A�μεσ’C�μέσα�μηδ'��A�μηδ'C�μήδε�μηδ’��A�μηδ’C�μήδε�μιλ.��A�μιλ.�μ’��A�μ’C�με�ν'��A�ν'C�να�ντ'��A�ντ'�ντ’��A�ντ’�ν’��A�ν’C�να�ολ'��A�ολ'C�όλος�ολ’��A�ολ’C�όλος�π.Χ.��A�π.Χ.�π.β.��A�π.β.�π.δ.��A�π.δ.�π.μ.��A�π.μ.�π.χ.��A�π.χ.�πάρ'��A�πάρ'C�παίρνω�πάρ’��A�πάρ’C�παίρνω�παρ'��A�παρ'C�παρά�παρ’��A�παρ’C�παρά�σ'��A�σ'C�σε�σ.��A�σ.�σ.α.λ.��A�σ.α.λ.�σ.σ.��A�σ.σ.�σελ.��A�σελ.�στρ.��A�στρ.�σ’��A�σ’C�σε�τ'��A�τ'C�το�τ'ς��A�τ'ς�τ.μ.��A�τ.μ.�τίποτ'��A�τίποτ'C�τίποτα�τίποτ’��A�τίποτ’C�τίποτα�τετ.��A�τετ.�τετρ.��A�τετρ.�τηλ.��A�τηλ.�τρισ.��A�τρισ.�τόν.��A�τόν.�τ’��A�τ’C�το�τ’ς��A�τ’ς�υπ'��A�υπ'C�υπό�υπ.��A�υπ.�υπ’��A�υπ’C�υπό�φέρ'��A�φέρ'C�φέρνω�φέρ’��A�φέρ’C�φέρνω�φερ'��A�φερ'C�φέρνω�φερ’��A�φερ’C�φέρνω�χ.μ.��A�χ.μ.�χγρ.��A�χγρ.�χιλ.��A�χιλ.�χλμ.��A�χλμ.�όμουρφ'��A�όμουρφ'C�όμορφος�όμουρφ’��A�όμουρφ’C�όμορφος�ύστερ'��A�ύστερ'C�ύστερα�ύστερ’��A�ύστερ’C�ύστερα�ಠ_ಠ��A�ಠ_ಠ�ಠ︵ಠ��A�ಠ︵ಠ�—��A�—�’��A�’�’ΣΟΥΝ��A�’ΣΟΥΝC�είμαι�’βγαινε��A�’βγαινεC�βγαίνω�’βλεπε��A�’βλεπεC�βλέπω�’γραφα��A�’γραφαC�γράφω�’γω��A�’γωC�εγώ�’δειξε��A�’δειξεC�δείχνω�’θελα��A�’θελαC�θέλω�’κανε��A�’κανεC�κάνω�’κει��A�’κειC�εκεί�’κλεβε��A�’κλεβεC�κλέβω�’λεγα��A�’λεγαC�λέγω�’λεγαν��A�’λεγανC�λέγω�’λεγε��A�’λεγεC�λέγω�’μας��A�’μαςC�εμάς�’μαστε��A�’μαστεC�είμαι�’μεις��A�’μειςC�εγώ�’μετ��A�’μετC�μετά�’μουν��A�’μουνC�είμαι�’μουνα��A�’μουναC�είμαι�’ναι��A�’ναιC�είμαι�’ξερες��A�’ξερεςC�ξέρω�’ξομολογήθηκε��A�’ξομολογήθηκεC�εξομολογούμαι�’παιρνα��A�’παιρναC�παίρνω�’πανε��A�’πανεC�λέγω�’πε��A�’πεC�λέγω�’πρεπε��A�’πρεπεC�πρέπει�’ρθε��A�’ρθεC�έρχομαι�’ρθεί��A�’ρθείC�έρχομαι�’ρθει��A�’ρθειC�έρχομαι�’ρθουν��A�’ρθουνC�έρχομαι�’ρθούνε��A�’ρθούνεC�έρχομαι�’ριξε��A�’ριξεC�ρίχνω�’ρχεται��A�’ρχεταιC�έρχομαι�’σένα��A�’σέναC�εγώ�’σβηνε��A�’σβηνεC�σβήνω�’σκασε��A�’σκασεC�σκάω�’σπαγαν��A�’σπαγανC�σπάω�’ταν��A�’τανC�είμαι�’τανε��A�’τανεC�είμαι�’φαγε��A�’φαγεC�τρώω�’φερε��A�’φερεC�φέρνω�’φύγε��A�’φύγεC�φεύγω�’χα��A�’χαC�έχω�’χαμε��A�’χαμεC�έχω�’χε��A�’χεC�έχω�’χει��A�’χειC�έχω�’χεις��A�’χειςC�έχω�’χουμε��A�’χουμεC�έχω�’χουν��A�’χουνC�έχω�’’��A�’’�faster_heuristics�
EL/el_cop_sigtyp_trf/trainable_lemmatizer/cfg ADDED
@@ -0,0 +1,479 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "labels":[
3
+ 8,
4
+ 10,
5
+ 11,
6
+ 13,
7
+ 16,
8
+ 20,
9
+ 22,
10
+ 25,
11
+ 28,
12
+ 29,
13
+ 30,
14
+ 31,
15
+ 35,
16
+ 36,
17
+ 39,
18
+ 43,
19
+ 48,
20
+ 49,
21
+ 50,
22
+ 52,
23
+ 54,
24
+ 56,
25
+ 58,
26
+ 62,
27
+ 66,
28
+ 67,
29
+ 70,
30
+ 74,
31
+ 78,
32
+ 81,
33
+ 82,
34
+ 87,
35
+ 91,
36
+ 95,
37
+ 96,
38
+ 99,
39
+ 100,
40
+ 103,
41
+ 106,
42
+ 109,
43
+ 110,
44
+ 112,
45
+ 115,
46
+ 117,
47
+ 121,
48
+ 123,
49
+ 125,
50
+ 127,
51
+ 130,
52
+ 135,
53
+ 138,
54
+ 140,
55
+ 144,
56
+ 147,
57
+ 150,
58
+ 153,
59
+ 156,
60
+ 159,
61
+ 162,
62
+ 167,
63
+ 168,
64
+ 170,
65
+ 174,
66
+ 176,
67
+ 177,
68
+ 182,
69
+ 184,
70
+ 185,
71
+ 188,
72
+ 193,
73
+ 198,
74
+ 203,
75
+ 205,
76
+ 207,
77
+ 209,
78
+ 214,
79
+ 216,
80
+ 218,
81
+ 219,
82
+ 225,
83
+ 228,
84
+ 229,
85
+ 233,
86
+ 235,
87
+ 236,
88
+ 240,
89
+ 241,
90
+ 246,
91
+ 248,
92
+ 252,
93
+ 257,
94
+ 261,
95
+ 265,
96
+ 268,
97
+ 270,
98
+ 272,
99
+ 274,
100
+ 275,
101
+ 277,
102
+ 278,
103
+ 280,
104
+ 282,
105
+ 285,
106
+ 286,
107
+ 288,
108
+ 289,
109
+ 290,
110
+ 292,
111
+ 296,
112
+ 297,
113
+ 299,
114
+ 300,
115
+ 302,
116
+ 303,
117
+ 306,
118
+ 311,
119
+ 313,
120
+ 315,
121
+ 318,
122
+ 321,
123
+ 323,
124
+ 326,
125
+ 327,
126
+ 330,
127
+ 331,
128
+ 338,
129
+ 342,
130
+ 344,
131
+ 345,
132
+ 347,
133
+ 348,
134
+ 351,
135
+ 355,
136
+ 356,
137
+ 359,
138
+ 361,
139
+ 362,
140
+ 365,
141
+ 367,
142
+ 369,
143
+ 374,
144
+ 378,
145
+ 379,
146
+ 385,
147
+ 387,
148
+ 390,
149
+ 391,
150
+ 393,
151
+ 396,
152
+ 397,
153
+ 401,
154
+ 404,
155
+ 408,
156
+ 412,
157
+ 413,
158
+ 416,
159
+ 418,
160
+ 420,
161
+ 426,
162
+ 428,
163
+ 429,
164
+ 431,
165
+ 433,
166
+ 435,
167
+ 439,
168
+ 442,
169
+ 444,
170
+ 447,
171
+ 449,
172
+ 452,
173
+ 453,
174
+ 454,
175
+ 456,
176
+ 460,
177
+ 464,
178
+ 468,
179
+ 469,
180
+ 472,
181
+ 474,
182
+ 476,
183
+ 477,
184
+ 478,
185
+ 479,
186
+ 481,
187
+ 483,
188
+ 484,
189
+ 485,
190
+ 488,
191
+ 489,
192
+ 491,
193
+ 494,
194
+ 496,
195
+ 501,
196
+ 503,
197
+ 506,
198
+ 507,
199
+ 510,
200
+ 513,
201
+ 514,
202
+ 515,
203
+ 517,
204
+ 520,
205
+ 522,
206
+ 523,
207
+ 527,
208
+ 532,
209
+ 533,
210
+ 535,
211
+ 539,
212
+ 543,
213
+ 546,
214
+ 548,
215
+ 549,
216
+ 551,
217
+ 553,
218
+ 554,
219
+ 558,
220
+ 559,
221
+ 560,
222
+ 562,
223
+ 566,
224
+ 570,
225
+ 571,
226
+ 575,
227
+ 577,
228
+ 579,
229
+ 582,
230
+ 584,
231
+ 586,
232
+ 587,
233
+ 588,
234
+ 589,
235
+ 592,
236
+ 594,
237
+ 597,
238
+ 600,
239
+ 601,
240
+ 603,
241
+ 604,
242
+ 608,
243
+ 611,
244
+ 614,
245
+ 615,
246
+ 616,
247
+ 618,
248
+ 620,
249
+ 623,
250
+ 627,
251
+ 629,
252
+ 630,
253
+ 631,
254
+ 634,
255
+ 637,
256
+ 638,
257
+ 640,
258
+ 645,
259
+ 650,
260
+ 652,
261
+ 654,
262
+ 655,
263
+ 660,
264
+ 663,
265
+ 669,
266
+ 671,
267
+ 674,
268
+ 675,
269
+ 680,
270
+ 682,
271
+ 684,
272
+ 688,
273
+ 690,
274
+ 691,
275
+ 692,
276
+ 696,
277
+ 697,
278
+ 699,
279
+ 701,
280
+ 706,
281
+ 710,
282
+ 712,
283
+ 713,
284
+ 715,
285
+ 716,
286
+ 717,
287
+ 719,
288
+ 721,
289
+ 722,
290
+ 726,
291
+ 727,
292
+ 729,
293
+ 731,
294
+ 733,
295
+ 734,
296
+ 735,
297
+ 737,
298
+ 738,
299
+ 740,
300
+ 742,
301
+ 745,
302
+ 749,
303
+ 753,
304
+ 756,
305
+ 758,
306
+ 759,
307
+ 762,
308
+ 767,
309
+ 768,
310
+ 770,
311
+ 771,
312
+ 772,
313
+ 777,
314
+ 778,
315
+ 781,
316
+ 782,
317
+ 784,
318
+ 785,
319
+ 787,
320
+ 788,
321
+ 792,
322
+ 794,
323
+ 795,
324
+ 796,
325
+ 798,
326
+ 801,
327
+ 802,
328
+ 803,
329
+ 804,
330
+ 809,
331
+ 814,
332
+ 816,
333
+ 817,
334
+ 819,
335
+ 820,
336
+ 822,
337
+ 826,
338
+ 829,
339
+ 830,
340
+ 832,
341
+ 836,
342
+ 839,
343
+ 844,
344
+ 846,
345
+ 847,
346
+ 850,
347
+ 853,
348
+ 854,
349
+ 856,
350
+ 858,
351
+ 861,
352
+ 863,
353
+ 865,
354
+ 868,
355
+ 870,
356
+ 872,
357
+ 876,
358
+ 878,
359
+ 881,
360
+ 884,
361
+ 887,
362
+ 888,
363
+ 891,
364
+ 892,
365
+ 895,
366
+ 899,
367
+ 900,
368
+ 903,
369
+ 906,
370
+ 908,
371
+ 909,
372
+ 910,
373
+ 912,
374
+ 913,
375
+ 917,
376
+ 918,
377
+ 921,
378
+ 922,
379
+ 926,
380
+ 929,
381
+ 932,
382
+ 934,
383
+ 937,
384
+ 939,
385
+ 940,
386
+ 942,
387
+ 944,
388
+ 946,
389
+ 947,
390
+ 951,
391
+ 952,
392
+ 955,
393
+ 957,
394
+ 958,
395
+ 960,
396
+ 964,
397
+ 968,
398
+ 969,
399
+ 970,
400
+ 972,
401
+ 977,
402
+ 979,
403
+ 980,
404
+ 981,
405
+ 984,
406
+ 985,
407
+ 986,
408
+ 987,
409
+ 988,
410
+ 989,
411
+ 991,
412
+ 992,
413
+ 994,
414
+ 998,
415
+ 999,
416
+ 1003,
417
+ 1006,
418
+ 1009,
419
+ 1011,
420
+ 1014,
421
+ 1017,
422
+ 1020,
423
+ 1021,
424
+ 1022,
425
+ 1024,
426
+ 1025,
427
+ 1027,
428
+ 1032,
429
+ 1035,
430
+ 1039,
431
+ 1041,
432
+ 1042,
433
+ 1045,
434
+ 1047,
435
+ 1048,
436
+ 1051,
437
+ 1052,
438
+ 1056,
439
+ 1058,
440
+ 1061,
441
+ 1062,
442
+ 1063,
443
+ 1066,
444
+ 1067,
445
+ 1068,
446
+ 1069,
447
+ 1073,
448
+ 1075,
449
+ 1079,
450
+ 1082,
451
+ 1083,
452
+ 1086,
453
+ 1087,
454
+ 1088,
455
+ 1089,
456
+ 1092,
457
+ 1094,
458
+ 1099,
459
+ 1102,
460
+ 1104,
461
+ 1105,
462
+ 1106,
463
+ 1108,
464
+ 1109,
465
+ 1112,
466
+ 1114,
467
+ 1117,
468
+ 1121,
469
+ 1124,
470
+ 1126,
471
+ 1128,
472
+ 1133,
473
+ 1135,
474
+ 1137,
475
+ 1138,
476
+ 1141,
477
+ 1142
478
+ ]
479
+ }
EL/el_cop_sigtyp_trf/trainable_lemmatizer/model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1297a44b325cc6c68ee4c97706e5e7ec15640253bdea6c22086597329e739355
3
+ size 1461753
EL/el_cop_sigtyp_trf/trainable_lemmatizer/trees ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:671198c541dcc7ce47a36c058e81dc63b768d9af65c4dbe2a01ffad572ed59b5
3
+ size 301923
EL/el_cop_sigtyp_trf/transformer/cfg ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "max_batch_items":4096
3
+ }
EL/el_cop_sigtyp_trf/transformer/model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2b67a310b5ff0be8ff7f5d94b7dbd737d271d4e3678a8fb4ca05825c8bdfde1
3
+ size 508592662
EL/el_cop_sigtyp_trf/vocab/key2row ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76be8b528d0075f7aae98d6fa57a6d3c83ae480a8469e668d7b0af968995ac71
3
+ size 1
EL/el_cop_sigtyp_trf/vocab/lookups.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76be8b528d0075f7aae98d6fa57a6d3c83ae480a8469e668d7b0af968995ac71
3
+ size 1
EL/el_cop_sigtyp_trf/vocab/strings.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:313a9305aa167e262271f555bc2076cc920da966ff9f40e2970d6832328e5292
3
+ size 1914563