Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: cop
|
| 3 |
+
widget:
|
| 4 |
+
- text: ⲁⲗⲗⲁ ⲁⲛⲟⲕ ⲁⲓⲥⲉⲧⲡⲧⲏⲩⲧⲛ ·
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
This is a [MicroBERT](https://github.com/lgessler/microbert) model for Coptic.
|
| 8 |
+
|
| 9 |
+
* Its suffix is **-m**, which means that it was pretrained using supervision from masked language modeling.
|
| 10 |
+
* The unlabeled Coptic data was taken from version 4.2.0 of the [Coptic SCRIPTORIUM corpus](https://github.com/copticscriptorium/corpora), totaling 970,642 tokens.
|
| 11 |
+
* The UD treebank [UD_Coptic_Scriptorium](https://github.com/UniversalDependencies/UD_Coptic-Scriptorium), v2.9, totaling 48,632 tokens, was used for labeled data.
|
| 12 |
+
|
| 13 |
+
Please see [the repository](https://github.com/lgessler/microbert) and
|
| 14 |
+
[the paper](https://github.com/lgessler/microbert/raw/master/MicroBERT__MRL_2022_.pdf) for more details.
|