geletka commited on
Commit
0c21df7
1 Parent(s): 32f6bff

Upload tokenizer_config.json

Browse files
Files changed (1) hide show
  1. tokenizer_config.json +37 -0
tokenizer_config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": null,
3
+ "cls_token": "[CLS]",
4
+ "cls_token_box": [
5
+ 0,
6
+ 0,
7
+ 0,
8
+ 0
9
+ ],
10
+ "do_basic_tokenize": true,
11
+ "do_lower_case": true,
12
+ "mask_token": "[MASK]",
13
+ "model_max_length": 512,
14
+ "name_or_path": "microsoft/layoutlmv2-base-uncased",
15
+ "never_split": null,
16
+ "only_label_first_subword": true,
17
+ "pad_token": "[PAD]",
18
+ "pad_token_box": [
19
+ 0,
20
+ 0,
21
+ 0,
22
+ 0
23
+ ],
24
+ "pad_token_label": -100,
25
+ "sep_token": "[SEP]",
26
+ "sep_token_box": [
27
+ 1000,
28
+ 1000,
29
+ 1000,
30
+ 1000
31
+ ],
32
+ "special_tokens_map_file": null,
33
+ "strip_accents": null,
34
+ "tokenize_chinese_chars": true,
35
+ "tokenizer_class": "LayoutLMv2Tokenizer",
36
+ "unk_token": "[UNK]"
37
+ }