add tokenizer
Browse files- special_tokens_map.json +5 -0
- spiece.model +3 -0
- tokenizer_config.json +12 -0
special_tokens_map.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"eos_token": "</s>",
|
3 |
+
"pad_token": "<pad>",
|
4 |
+
"unk_token": "<unk>"
|
5 |
+
}
|
spiece.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2dabb5302a55dfe096bdb96cee79e2aa5676a9d158ff6ae8e034f064191eac0a
|
3 |
+
size 916494
|
tokenizer_config.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": null,
|
3 |
+
"eos_token": "</s>",
|
4 |
+
"extra_ids": 0,
|
5 |
+
"name_or_path": "/content/drive/MyDrive/Finetune/urt5-base-news",
|
6 |
+
"pad_token": "<pad>",
|
7 |
+
"padding_side": "left",
|
8 |
+
"sp_model_kwargs": {},
|
9 |
+
"special_tokens_map_file": "/content/drive/MyDrive/urt5-base/special_tokens_map.json",
|
10 |
+
"tokenizer_class": "T5Tokenizer",
|
11 |
+
"unk_token": "<unk>"
|
12 |
+
}
|