#2: Finetune 10k 30 epochs
Browse files
README.md
CHANGED
|
@@ -2,7 +2,10 @@
|
|
| 2 |
license: mit
|
| 3 |
base_model: VietAI/envit5-base
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
model-index:
|
| 7 |
- name: envit5-base-iwslt15
|
| 8 |
results: []
|
|
@@ -14,6 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
# envit5-base-iwslt15
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [VietAI/envit5-base](https://huggingface.co/VietAI/envit5-base) on an unknown dataset.
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
@@ -42,6 +48,22 @@ The following hyperparameters were used during training:
|
|
| 42 |
- num_epochs: 10
|
| 43 |
- mixed_precision_training: Native AMP
|
| 44 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
### Framework versions
|
| 46 |
|
| 47 |
- Transformers 4.41.2
|
|
|
|
| 2 |
license: mit
|
| 3 |
base_model: VietAI/envit5-base
|
| 4 |
tags:
|
| 5 |
+
- translation
|
| 6 |
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- bleu
|
| 9 |
model-index:
|
| 10 |
- name: envit5-base-iwslt15
|
| 11 |
results: []
|
|
|
|
| 17 |
# envit5-base-iwslt15
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [VietAI/envit5-base](https://huggingface.co/VietAI/envit5-base) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 2.2687
|
| 22 |
+
- Bleu: 21.8184
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 48 |
- num_epochs: 10
|
| 49 |
- mixed_precision_training: Native AMP
|
| 50 |
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Bleu |
|
| 54 |
+
|:-------------:|:-----:|:-----:|:---------------:|:-------:|
|
| 55 |
+
| 2.0209 | 1.0 | 1250 | 1.7844 | 20.7717 |
|
| 56 |
+
| 1.5711 | 2.0 | 2500 | 1.7072 | 22.0149 |
|
| 57 |
+
| 1.2667 | 3.0 | 3750 | 1.7304 | 22.3730 |
|
| 58 |
+
| 1.0436 | 4.0 | 5000 | 1.7903 | 22.0901 |
|
| 59 |
+
| 0.8655 | 5.0 | 6250 | 1.8831 | 22.0823 |
|
| 60 |
+
| 0.7478 | 6.0 | 7500 | 1.9738 | 22.0309 |
|
| 61 |
+
| 0.6292 | 7.0 | 8750 | 2.0935 | 21.9696 |
|
| 62 |
+
| 0.5586 | 8.0 | 10000 | 2.1611 | 22.1045 |
|
| 63 |
+
| 0.5046 | 9.0 | 11250 | 2.2271 | 21.7866 |
|
| 64 |
+
| 0.4626 | 10.0 | 12500 | 2.2687 | 21.8184 |
|
| 65 |
+
|
| 66 |
+
|
| 67 |
### Framework versions
|
| 68 |
|
| 69 |
- Transformers 4.41.2
|