|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- bleu |
|
model-index: |
|
- name: opus-mt-mul-en-finetuned-lfn-to-en |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# opus-mt-mul-en-finetuned-lfn-to-en |
|
|
|
This model is a fine-tuned version of [Helsinki-NLP/opus-mt-mul-en](https://huggingface.co/Helsinki-NLP/opus-mt-mul-en) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.6492 |
|
- Bleu: 60.0647 |
|
- Gen Len: 8.5545 |
|
|
|
## Model description |
|
|
|
More information needed |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
More information needed |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 2e-05 |
|
- train_batch_size: 64 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 10 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:| |
|
| No log | 1.0 | 145 | 0.7224 | 55.1224 | 8.4971 | |
|
| No log | 2.0 | 290 | 0.6699 | 58.0064 | 8.5156 | |
|
| No log | 3.0 | 435 | 0.6526 | 59.1039 | 8.5165 | |
|
| 0.6334 | 4.0 | 580 | 0.6482 | 59.0726 | 8.5973 | |
|
| 0.6334 | 5.0 | 725 | 0.6454 | 59.0775 | 8.5593 | |
|
| 0.6334 | 6.0 | 870 | 0.6450 | 59.4424 | 8.5535 | |
|
| 0.3496 | 7.0 | 1015 | 0.6452 | 60.0632 | 8.5311 | |
|
| 0.3496 | 8.0 | 1160 | 0.6477 | 60.2116 | 8.5593 | |
|
| 0.3496 | 9.0 | 1305 | 0.6489 | 59.9325 | 8.5545 | |
|
| 0.3496 | 10.0 | 1450 | 0.6492 | 60.0647 | 8.5545 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.27.3 |
|
- Pytorch 1.13.1+cu116 |
|
- Datasets 2.10.1 |
|
- Tokenizers 0.13.2 |
|
|
