Upload processor
Browse files- special_tokens_map.json +1 -0
- tokenizer.json +9 -0
- tokenizer_config.json +1 -1
special_tokens_map.json
CHANGED
@@ -2,5 +2,6 @@
|
|
2 |
"bos_token": "<s>",
|
3 |
"eos_token": "</s>",
|
4 |
"pad_token": "<pad>",
|
|
|
5 |
"unk_token": "<unk>"
|
6 |
}
|
|
|
2 |
"bos_token": "<s>",
|
3 |
"eos_token": "</s>",
|
4 |
"pad_token": "<pad>",
|
5 |
+
"sep_token": "<sep/>",
|
6 |
"unk_token": "<unk>"
|
7 |
}
|
tokenizer.json
CHANGED
@@ -200,6 +200,15 @@
|
|
200 |
"rstrip": false,
|
201 |
"normalized": true,
|
202 |
"special": false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
203 |
}
|
204 |
],
|
205 |
"normalizer": {
|
|
|
200 |
"rstrip": false,
|
201 |
"normalized": true,
|
202 |
"special": false
|
203 |
+
},
|
204 |
+
{
|
205 |
+
"id": 75,
|
206 |
+
"content": "<sep/>",
|
207 |
+
"single_word": false,
|
208 |
+
"lstrip": false,
|
209 |
+
"rstrip": false,
|
210 |
+
"normalized": false,
|
211 |
+
"special": true
|
212 |
}
|
213 |
],
|
214 |
"normalizer": {
|
tokenizer_config.json
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
{
|
2 |
"model_max_length": 1000000000000000019884624838656,
|
3 |
-
"name_or_path": "./
|
4 |
"processor_class": "DonutProcessor",
|
5 |
"special_tokens_map_file": "./model_char_1/special_tokens_map.json",
|
6 |
"tokenizer_class": "PreTrainedTokenizerFast"
|
|
|
1 |
{
|
2 |
"model_max_length": 1000000000000000019884624838656,
|
3 |
+
"name_or_path": "./model_char_1_fix",
|
4 |
"processor_class": "DonutProcessor",
|
5 |
"special_tokens_map_file": "./model_char_1/special_tokens_map.json",
|
6 |
"tokenizer_class": "PreTrainedTokenizerFast"
|