Commit
•
3194193
1
Parent(s):
a5dfcaf
Upload tokenizer
Browse files- README.md +1 -1
- added_tokens.json +0 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +20 -0
- tokenizer.json +0 -0
- tokenizer_config.json +0 -0
README.md
CHANGED
|
@@ -2,4 +2,4 @@
|
|
| 2 |
metrics:
|
| 3 |
- cer
|
| 4 |
pipeline_tag: image-to-text
|
| 5 |
-
---
|
|
|
|
| 2 |
metrics:
|
| 3 |
- cer
|
| 4 |
pipeline_tag: image-to-text
|
| 5 |
+
---
|
added_tokens.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49c4ba4e495ddf31eb2fdba7fc6aef3c233091d25d35bc9d24694ccf48ae114c
|
| 3 |
+
size 904693
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<s>NOTUSED",
|
| 4 |
+
"</s>NOTUSED",
|
| 5 |
+
"<_>"
|
| 6 |
+
],
|
| 7 |
+
"bos_token": "<s>",
|
| 8 |
+
"cls_token": "<s>",
|
| 9 |
+
"eos_token": "</s>",
|
| 10 |
+
"mask_token": {
|
| 11 |
+
"content": "<mask>",
|
| 12 |
+
"lstrip": true,
|
| 13 |
+
"normalized": false,
|
| 14 |
+
"rstrip": false,
|
| 15 |
+
"single_word": false
|
| 16 |
+
},
|
| 17 |
+
"pad_token": "<pad>",
|
| 18 |
+
"sep_token": "</s>",
|
| 19 |
+
"unk_token": "<unk>"
|
| 20 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|