Fix link to herbert.
Browse files
README.md
CHANGED
|
@@ -11,7 +11,7 @@ The model was created for selective question answering in Polish. I.e. it is use
|
|
| 11 |
It is used to encode the contexts (aka passages) in the DPR bi-encoder architecture. The architecture requires two separate models.
|
| 12 |
The question part has to be encoded with the corresponding [question encoder](https://huggingface.co/enelpol/czywiesz-question).
|
| 13 |
|
| 14 |
-
The model was created by fine-tuning [Herbert base](https://huggingface.co/allegro/herbert-base) on "Czywiesz" dataset.
|
| 15 |
[Czywiesz](https://clarin-pl.eu/dspace/handle/11321/39) dataset contains questions and Wikipedia articles extracted from the Polish Wikipedia.
|
| 16 |
|
| 17 |
|
|
|
|
| 11 |
It is used to encode the contexts (aka passages) in the DPR bi-encoder architecture. The architecture requires two separate models.
|
| 12 |
The question part has to be encoded with the corresponding [question encoder](https://huggingface.co/enelpol/czywiesz-question).
|
| 13 |
|
| 14 |
+
The model was created by fine-tuning [Herbert base cased](https://huggingface.co/allegro/herbert-base-cased) on "Czywiesz" dataset.
|
| 15 |
[Czywiesz](https://clarin-pl.eu/dspace/handle/11321/39) dataset contains questions and Wikipedia articles extracted from the Polish Wikipedia.
|
| 16 |
|
| 17 |
|