Instructions to use utter-project/hutter-12-3rd-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Fairseq
How to use utter-project/hutter-12-3rd-base with Fairseq:
from fairseq.checkpoint_utils import load_model_ensemble_and_task_from_hf_hub models, cfg, task = load_model_ensemble_and_task_from_hf_hub( "utter-project/hutter-12-3rd-base" ) - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -16,4 +16,10 @@ language:
|
|
| 16 |
- sw
|
| 17 |
- gn
|
| 18 |
library_name: fairseq
|
| 19 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
- sw
|
| 17 |
- gn
|
| 18 |
library_name: fairseq
|
| 19 |
+
---
|
| 20 |
+
|
| 21 |
+
HUTTER-12: H(uBERT) UTTER model covering 12 languages:
|
| 22 |
+
Total training hours: 1,622
|
| 23 |
+
Number of updates: 400k
|
| 24 |
+
Number of iterations: 3
|
| 25 |
+
Clustering approach: mini-batch K-means (100% of the data)
|