add tokenizer info
Browse files- special_tokens_map.json +9 -0
- tokenizer_config.json +12 -0
special_tokens_map.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[CLS]",
|
| 3 |
+
"eos_token": "[SEP]",
|
| 4 |
+
"unk_token": "[UNK]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"pad_token": "[PAD]",
|
| 7 |
+
"cls_token": "[CLS]",
|
| 8 |
+
"mask_token": "[MASK]"
|
| 9 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_lower_case": true,
|
| 3 |
+
"unk_token": "[UNK]",
|
| 4 |
+
"sep_token": "[SEP]",
|
| 5 |
+
"pad_token": "[PAD]",
|
| 6 |
+
"cls_token": "[CLS]",
|
| 7 |
+
"mask_token": "[MASK]",
|
| 8 |
+
"tokenize_chinese_chars": true,
|
| 9 |
+
"strip_accents": null,
|
| 10 |
+
"model_max_length": 512,
|
| 11 |
+
"name_or_path": "HooshvareLab/bert-fa-base-uncased"
|
| 12 |
+
}
|