File size: 875 Bytes
badbda5 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 |
---
language:
- da
license: apache-2.0
---
## daT5-large
A smaller version of [Google's mt5-large](https://huggingface.co/google/mt5-base) model, where the original model is reduced to only include Danish embeddings.
## How to use
```python
from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("emillykkejensen/daT5-large")
model = AutoModel.from_pretrained("emillykkejensen/daT5-large")
```
## Further reading
[Gist](https://gist.github.com/emillykkejensen/8bf1b323495efc7252dee966e6bc1b5c) showing (in Danish) how the embeddings are extracted (for mt5-base)
[Article](https://towardsdatascience.com/how-to-adapt-a-multilingual-t5-model-for-a-single-language-b9f94f3d9c90) explaining how to do it by [David Dale](https://huggingface.co/cointegrated)
## Also check out
[daT5-base](https://huggingface.co/emillykkejensen/daT5-base) |