Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -16,4 +16,19 @@ configs:
|
|
| 16 |
data_files:
|
| 17 |
- split: train
|
| 18 |
path: data/train-*
|
|
|
|
|
|
|
|
|
|
| 19 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
data_files:
|
| 17 |
- split: train
|
| 18 |
path: data/train-*
|
| 19 |
+
license: cc-by-sa-3.0
|
| 20 |
+
language:
|
| 21 |
+
- ga
|
| 22 |
---
|
| 23 |
+
|
| 24 |
+
# Irish Wikipedia - Vicipéid (Gaeilge)
|
| 25 |
+
This dataset is generated from the [20240501 wikimedia dump](https://dumps.wikimedia.org/gawiki/20240501/).
|
| 26 |
+
|
| 27 |
+
The "text" column is obtained by combining the article_title and article_text from each wikipedia article, as follows:
|
| 28 |
+
```
|
| 29 |
+
text = f'{article_title}. {article_text}'
|
| 30 |
+
```
|
| 31 |
+
Token_count is obtained by using the Llama 3 tokenizer.
|
| 32 |
+
|
| 33 |
+
# License
|
| 34 |
+
This dataset is derived from content from Wikipedia, licensed under CC BY-SA 3.0. The original content is available at [https://ga.wikipedia.org/]().
|