Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,40 @@
|
|
1 |
---
|
|
|
|
|
|
|
|
|
2 |
license: apache-2.0
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- hu
|
4 |
+
tags:
|
5 |
+
- OCR cleaning
|
6 |
license: apache-2.0
|
7 |
+
widget:
|
8 |
+
- text: >-
|
9 |
+
trans: Bálvándy szokott kedves modoraval mulattata Dorothea ;rófnőt.
|
10 |
---
|
11 |
+
|
12 |
+
# mT5 Translation model
|
13 |
+
|
14 |
+
For further models, scripts and details, see or [our demo site](https://juniper.nytud.hu/demo/nlp).
|
15 |
+
|
16 |
+
- Pretrained model used: mT5-base
|
17 |
+
- Prefix: "trans: "
|
18 |
+
|
19 |
+
## Limitations
|
20 |
+
|
21 |
+
- max_source_length = 256
|
22 |
+
- max_target_length = 256
|
23 |
+
|
24 |
+
|
25 |
+
## Citation
|
26 |
+
If you use this model, please cite the following paper:
|
27 |
+
|
28 |
+
```
|
29 |
+
|
30 |
+
@inproceedings {laki-ocr,
|
31 |
+
title = {OCR hibák javítása neurális technológiák segítségével},
|
32 |
+
booktitle = {XVIII. Magyar Számítógépes Nyelvészeti Konferencia},
|
33 |
+
year = {2022},
|
34 |
+
publisher = {Szegedi Tudományegyetem, Informatikai Intézet},
|
35 |
+
address = {Szeged, Magyarország},
|
36 |
+
author = {Laki, László and and Kőrös, Ádám and Ligeti-Nagy, Noémi and and Nyéki, Bence and Vadász, Noémi and Yang, Zijian Győző and Váradi Tamás},
|
37 |
+
pages = {417--430}
|
38 |
+
}
|
39 |
+
|
40 |
+
```
|