Downloads · 30 days
38
36% of all-time downloads
clarin-pl/combo-nlp-xlm-roberta-base-persian-perdt-ud2.17
combo-nlp-xlm-roberta-base-persian-perdt-ud2.17 is a token classification model from clarin-pl. Use it when you need labels on individual words, such as names. The card lists the license as cc-by-sa-4.0.
This is a Persian-language model based on combo-nlp, an open-source natural language preprocessing system. It performs:
Downloads · 30 days
38
36% of all-time downloads
All-time downloads
105
Public
Repo size
1.1 GB
Likes
0
Public
Click a slice to open those files.
.bin1.1 GB · 100%
From the Hugging Face model README
This is a Persian-language model based on combo-nlp, an open-source natural language preprocessing system. It performs:
The Persian model uses FacebookAI/xlm-roberta-base as its base encoder and is trained on UD_Persian-PerDT (UD v2.17).
Evaluation was performed on the UD_Persian-PerDT test split using the standard CoNLL 2018 eval script.
Two evaluation rows are reported:
| Metric | Tokens | Sentences | Words | UPOS | XPOS | UFeats | AllTags | Lemmas |
|---|---|---|---|---|---|---|---|---|
| Full-text (F1) | 99.85 | 99.82 | 99.25 | 97.26 | 97.24 | 97.36 | 95.50 | 98.73 |
| Aligned accuracy | 0.00 | 0.00 | 0.00 | 98.00 | 97.97 | 98.10 | 96.22 | 99.47 |
| Metric | UAS | LAS | CLAS | MLAS | BLEX |
|---|---|---|---|---|---|
| Full-text (F1) | 94.18 | 92.26 | 90.59 | 87.17 | 90.00 |
| Aligned accuracy | 94.89 | 92.96 | 91.55 | 88.10 | 90.96 |
Install the library from PyPI (assuming you have a virtual environment created):
pip install combo-nlp
The combo-seg segmenter (used to split and
tokenise raw text) is installed automatically as a dependency of combo-nlp, so
no extra install step is needed.
from combo import COMBO
# Load a pre-trained model with the corresponding combo-seg segmenter
nlp = COMBO("Persian")
# Parse raw text (handles sentence splitting + tokenization)
result = nlp("روباه قهوهای چابک از روی سگ تنبل میپرد.")
# Inspect results
for sentence in result:
for token in sentence:
print(f"{token.form:<15} {token.lemma:<15} {token.upos:<8} head={token.head} {token.deprel}")
Refer to the combo-nlp documentation for installation and usage instructions:
The training data license: cc-by-sa-4.0 is derived from the Universal Dependencies treebank. For the full license terms, please refer to the LICENSE.txt file in the treebank repository:
If you use this model, please cite:
Ulewicz, M., Jabłońska, M., Klimaszewski, M., Przybyła, P., Pszenny, Ł., Rybak, P., Wiącek, M., & Wróblewska, A. (2026). COMBO-NLP Models Trained on UD v2.17. Zenodo. https://doi.org/10.5281/zenodo.19650523
@software{combo_nlp_2026,
author = {Ulewicz, Michał and Jabłońska, Maja and Klimaszewski, Mateusz and Przybyła, Piotr and Pszenny, Łukasz and Rybak, Piotr and Wiącek, Martyna and Wróblewska, Alina},
title = {{COMBO-NLP} Models Trained on {UD} v2.17},
year = {2026},
publisher = {Zenodo},
doi = {10.5281/zenodo.19650523},
url = {https://doi.org/10.5281/zenodo.19650523}
}
We gratefully acknowledge Polish high-performance computing infrastructure PLGrid (HPC Center: ACK Cyfronet AGH) for providing computer facilities and support within computational grant no. PLG/2025/018815.