lgessler's picture
Create README.md
3982b8f
|
raw
history blame contribute delete
No virus
752 Bytes
---
language: mlt
widget:
- text: Hu rrifjuta l-akkużi li l-Gvern Malti qed jimxi bi djufija.
---
This is a [MicroBERT](https://github.com/lgessler/microbert) model for Maltese.
* Its suffix is **-mx**, which means that it was pretrained using supervision from masked language modeling and XPOS tagging.
* The unlabeled Maltese data was taken from a February 2022 dump of Maltese Wikipedia, totaling 2,113,223 tokens.
* The UD treebank [UD_Maltese-GSD](https://github.com/UniversalDependencies/UD_Maltese-MUDT), v2.9, totaling 44,162 tokens, was used for labeled data.
Please see [the repository](https://github.com/lgessler/microbert) and
[the paper](https://github.com/lgessler/microbert/raw/master/MicroBERT__MRL_2022_.pdf) for more details.