metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: opus-mt-en-mul-finetuned-lfn-to-en
results: []
opus-mt-en-mul-finetuned-lfn-to-en
This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-mul on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6208
- Bleu: 62.9717
- Gen Len: 11.5165
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: 32
- eval_batch_size: 32
- 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 | 290 | 0.9193 | 51.7454 | 11.5088 |
| 1.2009 | 2.0 | 580 | 0.7614 | 56.6817 | 11.4971 |
| 1.2009 | 3.0 | 870 | 0.6865 | 59.4575 | 11.4815 |
| 0.6524 | 4.0 | 1160 | 0.6545 | 60.9631 | 11.5088 |
| 0.6524 | 5.0 | 1450 | 0.6360 | 61.8171 | 11.5039 |
| 0.4903 | 6.0 | 1740 | 0.6337 | 61.9929 | 11.5049 |
| 0.4064 | 7.0 | 2030 | 0.6269 | 62.8025 | 11.5146 |
| 0.4064 | 8.0 | 2320 | 0.6234 | 62.5979 | 11.5292 |
| 0.3434 | 9.0 | 2610 | 0.6197 | 63.0131 | 11.5428 |
| 0.3434 | 10.0 | 2900 | 0.6208 | 62.9717 | 11.5165 |
Framework versions
- Transformers 4.27.3
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2