Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
## Description
|
| 2 |
+
|
| 3 |
+
Afroscope is a MODEL_TYPE (e.g., FastText / Transformer) language identification (LID) model that supports more than N_LABELS labels.
|
| 4 |
+
|
| 5 |
+
**Latest:** MODEL_NAME is now updated to VERSION. VERSION supports N_LABELS labels (LABEL_FORMAT_HERE).
|
| 6 |
+
|
| 7 |
+
For more details on the supported languages and performance, as well as significant changes from previous versions, please refer to LINK_HERE.
|
| 8 |
+
|
| 9 |
+
- **Demo:** huggingface
|
| 10 |
+
- **Repository:** github
|
| 11 |
+
- **Paper:** paper (VENUE YEAR)
|
| 12 |
+
- **Point of Contact:** EMAIL@DOMAIN.COM
|
| 13 |
+
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
## How to use
|
| 17 |
+
|
| 18 |
+
Here is how to use this model to detect the language of a given text:
|
| 19 |
+
|
| 20 |
+
```python
|
| 21 |
+
import fasttext
|
| 22 |
+
from huggingface_hub import hf_hub_download
|
| 23 |
+
|
| 24 |
+
# model.bin is the latest version always
|
| 25 |
+
model_path = hf_hub_download(repo_id="ORG_OR_USER/MODEL_NAME", filename="model.bin")
|
| 26 |
+
|
| 27 |
+
model = fasttext.load_model(model_path)
|
| 28 |
+
model.predict("Hello, world!")
|
| 29 |
+
```
|
| 30 |
+
|
| 31 |
+
## Citation
|
| 32 |
+
|
| 33 |
+
```bibtex
|
| 34 |
+
@article{kwon2026afroscope,
|
| 35 |
+
title={AfroScope: A Framework for Studying the Linguistic Landscape of Africa},
|
| 36 |
+
author={Kwon, Sang Yun and Elmadany, AbdelRahim and Abdul-Mageed, Muhammad},
|
| 37 |
+
journal={arXiv preprint arXiv:2601.13346},
|
| 38 |
+
year={2026}
|
| 39 |
+
}
|
| 40 |
+
```
|