bernardocecchetto commited on
Commit
91c9eb4
·
verified ·
1 Parent(s): f799826

Upload tokenizer

Browse files
Files changed (3) hide show
  1. added_tokens.json +3 -0
  2. tokenizer.json +9 -0
  3. tokenizer_config.json +3 -0
added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "[ENDERECO URL]": 119547
3
+ }
tokenizer.json CHANGED
@@ -59,6 +59,15 @@
59
  "rstrip": false,
60
  "normalized": false,
61
  "special": true
 
 
 
 
 
 
 
 
 
62
  }
63
  ],
64
  "normalizer": {
 
59
  "rstrip": false,
60
  "normalized": false,
61
  "special": true
62
+ },
63
+ {
64
+ "id": 119547,
65
+ "content": "[ENDERECO URL]",
66
+ "single_word": false,
67
+ "lstrip": false,
68
+ "rstrip": false,
69
+ "normalized": true,
70
+ "special": false
71
  }
72
  ],
73
  "normalizer": {
tokenizer_config.json CHANGED
@@ -5,10 +5,13 @@
5
  "ignore_mismatched_sizes": true,
6
  "mask_token": "[MASK]",
7
  "model_max_length": 512,
 
8
  "pad_token": "[PAD]",
 
9
  "sep_token": "[SEP]",
10
  "strip_accents": null,
11
  "tokenize_chinese_chars": true,
12
  "tokenizer_class": "DistilBertTokenizer",
 
13
  "unk_token": "[UNK]"
14
  }
 
5
  "ignore_mismatched_sizes": true,
6
  "mask_token": "[MASK]",
7
  "model_max_length": 512,
8
+ "normalization": true,
9
  "pad_token": "[PAD]",
10
+ "padding": true,
11
  "sep_token": "[SEP]",
12
  "strip_accents": null,
13
  "tokenize_chinese_chars": true,
14
  "tokenizer_class": "DistilBertTokenizer",
15
+ "truncation": true,
16
  "unk_token": "[UNK]"
17
  }