Instructions to use rishi70612/nepali-tamang-english-mt with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Fairseq
How to use rishi70612/nepali-tamang-english-mt 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( "rishi70612/nepali-tamang-english-mt" ) - Notebooks
- Google Colab
- Kaggle
Update config.json with repetition penalties
Browse files- config.json +2 -2
config.json
CHANGED
|
@@ -34,6 +34,6 @@
|
|
| 34 |
"num_beams": 5,
|
| 35 |
"early_stopping": false,
|
| 36 |
"length_penalty": 1.0,
|
| 37 |
-
"no_repeat_ngram_size":
|
| 38 |
-
"repetition_penalty":
|
| 39 |
}
|
|
|
|
| 34 |
"num_beams": 5,
|
| 35 |
"early_stopping": false,
|
| 36 |
"length_penalty": 1.0,
|
| 37 |
+
"no_repeat_ngram_size": 2,
|
| 38 |
+
"repetition_penalty": 2.0
|
| 39 |
}
|