File size: 2,100 Bytes
23c3f2b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 |
---
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
|