| license: cc-by-sa-4.0 | |
| language: | |
| - et | |
| base_model: | |
| - EMBEDDIA/est-roberta | |
| pipeline_tag: token-classification | |
| library_name: transformers | |
| ## Model description | |
| est-roberta-vm-morph-tagging is an Est-RoBERTa based model fine-tuned for tagging partofpseech and morphological features in Estonian texts using [Vabamorf's tagset](https://github.com/Filosoft/vabamorf/blob/e6d42371006710175f7ec328c98f90b122930555/doc/tagset.md). | |
| This is version 2 of the model, which has been: 1) fine-tuned to mimic automatic tagging behaviour of the [Vabamorf morphological analyser and disambiguator](https://github.com/estnltk/estnltk/blob/main/tutorials/nlp_pipeline/B_morphology/01_morphological_analysis.ipynb) on an automatically tagged corpus, 2) fine-tuned on gold standard morphological annotations of Estonian Dependency Treebank (EDT). | |
| Fine-tuning and evaluation scripts of the model are available in the repository: https://github.com/estnltk/estnltk-model-training/tree/main/morph_tagging |