Flova commited on
Commit
32c0c95
1 Parent(s): 9148d32

Upload processor

Browse files
Files changed (3) hide show
  1. special_tokens_map.json +0 -13
  2. tokenizer.json +0 -0
  3. tokenizer_config.json +3 -18
special_tokens_map.json CHANGED
@@ -1,19 +1,6 @@
1
  {
2
- "additional_special_tokens": [
3
- "<s_iitcdip>",
4
- "<s_synthdog>"
5
- ],
6
  "bos_token": "<s>",
7
- "cls_token": "<s>",
8
  "eos_token": "</s>",
9
- "mask_token": {
10
- "content": "<mask>",
11
- "lstrip": true,
12
- "normalized": true,
13
- "rstrip": false,
14
- "single_word": false
15
- },
16
  "pad_token": "<pad>",
17
- "sep_token": "</s>",
18
  "unk_token": "<unk>"
19
  }
 
1
  {
 
 
 
 
2
  "bos_token": "<s>",
 
3
  "eos_token": "</s>",
 
 
 
 
 
 
 
4
  "pad_token": "<pad>",
 
5
  "unk_token": "<unk>"
6
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json CHANGED
@@ -1,22 +1,7 @@
1
  {
2
- "bos_token": "<s>",
3
- "cls_token": "<s>",
4
- "eos_token": "</s>",
5
- "mask_token": {
6
- "__type": "AddedToken",
7
- "content": "<mask>",
8
- "lstrip": true,
9
- "normalized": true,
10
- "rstrip": false,
11
- "single_word": false
12
- },
13
  "model_max_length": 1000000000000000019884624838656,
14
- "name_or_path": "./model_extended_2",
15
- "pad_token": "<pad>",
16
  "processor_class": "DonutProcessor",
17
- "sep_token": "</s>",
18
- "sp_model_kwargs": {},
19
- "special_tokens_map_file": null,
20
- "tokenizer_class": "XLMRobertaTokenizer",
21
- "unk_token": "<unk>"
22
  }
 
1
  {
 
 
 
 
 
 
 
 
 
 
 
2
  "model_max_length": 1000000000000000019884624838656,
3
+ "name_or_path": "./model_char_1/",
 
4
  "processor_class": "DonutProcessor",
5
+ "special_tokens_map_file": "./model_char_1/special_tokens_map.json",
6
+ "tokenizer_class": "PreTrainedTokenizerFast"
 
 
 
7
  }