DylanonWic
commited on
Commit
•
070459a
1
Parent(s):
a687122
Upload 4 files
Browse files- added_tokens.json +4 -0
- special_tokens_map.json +22 -0
- tokenizer_config.json +2 -3
added_tokens.json
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"</s>": 71,
|
3 |
+
"<s>": 70
|
4 |
+
}
|
special_tokens_map.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
{
|
4 |
+
"content": "<s>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": true,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false
|
9 |
+
},
|
10 |
+
{
|
11 |
+
"content": "</s>",
|
12 |
+
"lstrip": false,
|
13 |
+
"normalized": true,
|
14 |
+
"rstrip": false,
|
15 |
+
"single_word": false
|
16 |
+
}
|
17 |
+
],
|
18 |
+
"bos_token": "<s>",
|
19 |
+
"eos_token": "</s>",
|
20 |
+
"pad_token": "[PAD]",
|
21 |
+
"unk_token": "[UNK]"
|
22 |
+
}
|
tokenizer_config.json
CHANGED
@@ -3,12 +3,11 @@
|
|
3 |
"do_lower_case": false,
|
4 |
"eos_token": "</s>",
|
5 |
"model_max_length": 1000000000000000019884624838656,
|
6 |
-
"
|
7 |
-
"num_proc": 4,
|
8 |
"pad_token": "[PAD]",
|
9 |
"replace_word_delimiter_char": " ",
|
10 |
"special_tokens_map_file": "/root/.cache/huggingface/hub/models--airesearch--wav2vec2-large-xlsr-53-th/snapshots/3155938c549b23eee16b1d4b55dcb161b7fe4bcf/special_tokens_map.json",
|
11 |
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
12 |
"unk_token": "[UNK]",
|
13 |
"word_delimiter_token": "|"
|
14 |
-
}
|
|
|
3 |
"do_lower_case": false,
|
4 |
"eos_token": "</s>",
|
5 |
"model_max_length": 1000000000000000019884624838656,
|
6 |
+
"num_proc": 2,
|
|
|
7 |
"pad_token": "[PAD]",
|
8 |
"replace_word_delimiter_char": " ",
|
9 |
"special_tokens_map_file": "/root/.cache/huggingface/hub/models--airesearch--wav2vec2-large-xlsr-53-th/snapshots/3155938c549b23eee16b1d4b55dcb161b7fe4bcf/special_tokens_map.json",
|
10 |
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
11 |
"unk_token": "[UNK]",
|
12 |
"word_delimiter_token": "|"
|
13 |
+
}
|