Commit
•
25ff595
1
Parent(s):
162f41c
Update README.md
Browse files
README.md
CHANGED
|
@@ -11,43 +11,39 @@ pipeline_tag: translation
|
|
| 11 |
---
|
| 12 |
|
| 13 |
|
| 14 |
-
## Finetuning
|
| 15 |
|
| 16 |
-
|
| 17 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
| 21 |
facebook/mbart-large-50-many-to-many-mmt finetuned for translation task in Telugu language
|
| 22 |
|
| 23 |
-
|
| 24 |
## Training and evaluation data
|
| 25 |
|
| 26 |
-
ai4bharath/samanantar
|
| 27 |
|
| 28 |
-
|
| 29 |
-
## Training hyperparameters
|
| 30 |
|
| 31 |
The following hyperparameters were used during training:
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
total_train_batch_size: 8
|
| 36 |
-
|
| 37 |
-
num_epochs: 11
|
| 38 |
|
| 39 |
## Benchamark Evaluation
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
BLUE score on IN-22: 26.069430553765887
|
| 44 |
|
| 45 |
## Framework versions
|
| 46 |
|
| 47 |
-
Transformers 4.42.3
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
Datasets 2.20.0
|
| 52 |
-
|
| 53 |
-
Tokenizers 0.19.1
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
|
|
|
|
| 14 |
|
| 15 |
+
# Finetuning
|
| 16 |
|
| 17 |
+
This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on the samanantar dataset.
|
| 18 |
+
|
| 19 |
+
source group: English
|
| 20 |
+
|
| 21 |
+
target group: Telugu
|
| 22 |
+
|
| 23 |
+
model: transformer
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
| 27 |
facebook/mbart-large-50-many-to-many-mmt finetuned for translation task in Telugu language
|
| 28 |
|
|
|
|
| 29 |
## Training and evaluation data
|
| 30 |
|
| 31 |
+
ai4bharath/samanantar
|
| 32 |
|
| 33 |
+
### Training hyperparameters
|
|
|
|
| 34 |
|
| 35 |
The following hyperparameters were used during training:
|
| 36 |
+
- learning_rate: 2e-5
|
| 37 |
+
- total_train_batch_size: 8
|
| 38 |
+
- num_epochs: 1
|
|
|
|
|
|
|
|
|
|
| 39 |
|
| 40 |
## Benchamark Evaluation
|
| 41 |
+
-BLEU score on Tatoeba: 11.208466750961147
|
| 42 |
+
-BLUE score on IN-22: 26.069430553765887
|
|
|
|
|
|
|
| 43 |
|
| 44 |
## Framework versions
|
| 45 |
|
| 46 |
+
-Transformers 4.42.3
|
| 47 |
+
-Pytorch 2.1.2
|
| 48 |
+
-Datasets 2.20.0
|
| 49 |
+
-Tokenizers 0.19.1
|
|
|
|
|
|
|
|
|