Update README.md
Browse files
README.md
CHANGED
|
@@ -42,25 +42,13 @@ The following hyperparameters were used during training:
|
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
- num_epochs: 10
|
| 44 |
|
| 45 |
-
### Training results
|
| 46 |
-
|
| 47 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
| 48 |
-
|:-------------:|:-----:|:-----:|:---------------:|
|
| 49 |
-
| 2.7664 | 1.0 | 2234 | 2.5507 |
|
| 50 |
-
| 2.489 | 2.0 | 4468 | 2.3811 |
|
| 51 |
-
| 2.353 | 3.0 | 6702 | 2.2984 |
|
| 52 |
-
| 2.2309 | 4.0 | 8936 | 2.2492 |
|
| 53 |
-
| 2.1537 | 5.0 | 11170 | 2.0289 |
|
| 54 |
-
| 2.0825 | 6.0 | 13404 | 2.0217 |
|
| 55 |
-
| 2.0479 | 7.0 | 15638 | 2.0163 |
|
| 56 |
-
| 1.9948 | 8.0 | 17872 | 2.0128 |
|
| 57 |
-
| 1.9858 | 9.0 | 20106 | 2.0094 |
|
| 58 |
-
| 1.9751 | 10.0 | 22340 | 2.0078 |
|
| 59 |
-
|
| 60 |
-
|
| 61 |
### Framework versions
|
| 62 |
|
| 63 |
- Transformers 4.35.2
|
| 64 |
- Pytorch 2.1.1+cu121
|
| 65 |
- Datasets 2.14.5
|
| 66 |
- Tokenizers 0.15.1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
- num_epochs: 10
|
| 44 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
### Framework versions
|
| 46 |
|
| 47 |
- Transformers 4.35.2
|
| 48 |
- Pytorch 2.1.1+cu121
|
| 49 |
- Datasets 2.14.5
|
| 50 |
- Tokenizers 0.15.1
|
| 51 |
+
|
| 52 |
+
### Benchmarks
|
| 53 |
+
- BLEU: 0.3449933819584583
|
| 54 |
+
- CHRF: 66.67228299384574
|