added model card
Browse files
README.md
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Hugging Face's logo
|
| 2 |
+
---
|
| 3 |
+
language:
|
| 4 |
+
- om
|
| 5 |
+
- am
|
| 6 |
+
- rw
|
| 7 |
+
- rn
|
| 8 |
+
- ha
|
| 9 |
+
- ig
|
| 10 |
+
- pcm
|
| 11 |
+
- so
|
| 12 |
+
- sw
|
| 13 |
+
- ti
|
| 14 |
+
- yo
|
| 15 |
+
- multilingual
|
| 16 |
+
datasets:
|
| 17 |
+
|
| 18 |
+
---
|
| 19 |
+
# AfriBERTa_small
|
| 20 |
+
## Model description
|
| 21 |
+
AfriBERTa small is a pretrained multilingual language model with around 97 million parameters.
|
| 22 |
+
The model has 4 layers, 6 attention heads, 768 hidden units and 3072 feed forward size.
|
| 23 |
+
The model was pretrained on 11 African languages namely - Afaan Oromoo (also called Oromo), Amharic, Gahuza (a mixed language containing Kinyarwanda and Kirundi), Hausa, Igbo, Nigerian Pidgin, Somali, Swahili, Tigrinya and Yorùbá.
|
| 24 |
+
The model has been shown to obtain competitive downstream performances on text classification and Named Entity Recognition on several African languages, including those it was not pretrained on.
|
| 25 |
+
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
#### How to use
|
| 30 |
+
You can use this model with Transformers for any downstream task.
|
| 31 |
+
For example, assuming we want to finetune this model on a token classification task, we do the following:
|
| 32 |
+
|
| 33 |
+
```python
|
| 34 |
+
>>> from transformers import AutoTokenizer, AutoModelForTokenClassification
|
| 35 |
+
>>> tokenizer = AutoTokenizer.from_pretrained("castorini/afriberta_small")
|
| 36 |
+
>>> model = AutoModelForTokenClassification.from_pretrained("castorini/afriberta_small")
|
| 37 |
+
```
|
| 38 |
+
|
| 39 |
+
#### Limitations and bias
|
| 40 |
+
This model is possibly limited by its training dataset which are majorly obtained from news articles from a specific span of time.
|
| 41 |
+
Thus, it may not generalize well.
|
| 42 |
+
|
| 43 |
+
## Training data
|
| 44 |
+
The model was trained on an aggregation of datasets from the BBC news website and Common Crawl.
|
| 45 |
+
|
| 46 |
+
## Training procedure
|
| 47 |
+
For information on training procedures, please refer to the AfriBERTa [paper]() or [repository](https://github.com/keleog/afriberta)
|
| 48 |
+
|
| 49 |
+
### BibTeX entry and citation info
|
| 50 |
+
```
|
| 51 |
+
Kelechi Ogueji, Yuxin Zhu, Jimmy Lin.
|
| 52 |
+
Small Data? No Problem! Exploring the Viability of Pretrained Multilingual Language Models for Low-resourced Languages
|
| 53 |
+
Proceedings of the 1st workshop on Multilingual Representation Learning at EMNLP 2021
|
| 54 |
+
```
|
| 55 |
+
|
| 56 |
+
|