File size: 4,342 Bytes
8bfbc6f |
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 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 |
---
library_name: transformers
license: apache-2.0
base_model: Helsinki-NLP/opus-mt-es-es
tags:
- generated_from_trainer
model-index:
- name: 8_6kmslsamples
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. -->
# 8_6kmslsamples
This model is a fine-tuned version of [Helsinki-NLP/opus-mt-es-es](https://huggingface.co./Helsinki-NLP/opus-mt-es-es) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4729
- Bleu Msl: 77.3631
- Bleu Asl: 0
- Ter Msl: 11.9346
- Ter Asl: 100
## 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: 0.0001
- train_batch_size: 32
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 30
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu Msl | Bleu Asl | Ter Msl | Ter Asl |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:-------:|:-------:|
| No log | 1.0 | 217 | 0.2646 | 21.2021 | 84.6183 | 57.1279 | 8.2427 |
| No log | 2.0 | 434 | 0.1929 | 17.6495 | 84.9583 | 34.5912 | 7.4059 |
| 0.4655 | 3.0 | 651 | 0.1742 | 67.4481 | 89.3963 | 19.9161 | 5.7531 |
| 0.4655 | 4.0 | 868 | 0.1567 | 63.4679 | 86.5311 | 24.6855 | 7.8243 |
| 0.0791 | 5.0 | 1085 | 0.1559 | 63.5398 | 82.6386 | 21.5933 | 9.6234 |
| 0.0791 | 6.0 | 1302 | 0.1548 | 73.9641 | 86.8345 | 17.8721 | 8.1172 |
| 0.0421 | 7.0 | 1519 | 0.1606 | 70.5606 | 88.7113 | 16.7715 | 5.8368 |
| 0.0421 | 8.0 | 1736 | 0.1556 | 35.3257 | 86.5104 | 30.7652 | 7.4686 |
| 0.0421 | 9.0 | 1953 | 0.1564 | 66.0661 | 85.8235 | 22.6415 | 8.2636 |
| 0.0269 | 10.0 | 2170 | 0.1584 | 68.5372 | 88.0956 | 20.1782 | 6.7364 |
| 0.0269 | 11.0 | 2387 | 0.1641 | 64.1749 | 89.0674 | 22.8512 | 5.8996 |
| 0.0178 | 12.0 | 2604 | 0.1668 | 72.4073 | 88.8679 | 16.5094 | 6.0460 |
| 0.0178 | 13.0 | 2821 | 0.1663 | 63.9882 | 86.2327 | 18.7107 | 8.8285 |
| 0.0136 | 14.0 | 3038 | 0.1646 | 73.1200 | 88.9404 | 17.1908 | 6.2343 |
| 0.0136 | 15.0 | 3255 | 0.1726 | 66.6138 | 87.2488 | 18.6583 | 6.9874 |
| 0.0136 | 16.0 | 3472 | 0.1716 | 54.0153 | 87.6643 | 22.4843 | 6.4644 |
| 0.0101 | 17.0 | 3689 | 0.1680 | 65.4861 | 88.5112 | 16.5618 | 5.7531 |
| 0.0101 | 18.0 | 3906 | 0.1733 | 72.5027 | 88.9905 | 17.1384 | 5.8787 |
| 0.0078 | 19.0 | 4123 | 0.1701 | 69.9889 | 85.8965 | 18.7631 | 8.7029 |
| 0.0078 | 20.0 | 4340 | 0.1775 | 69.8437 | 88.9591 | 17.6101 | 6.1297 |
| 0.0051 | 21.0 | 4557 | 0.1731 | 72.1034 | 88.3815 | 18.0818 | 6.5900 |
| 0.0051 | 22.0 | 4774 | 0.1744 | 69.9773 | 89.2154 | 17.0860 | 5.7531 |
| 0.0051 | 23.0 | 4991 | 0.1711 | 72.3725 | 90.1699 | 15.6709 | 5.3975 |
| 0.0059 | 24.0 | 5208 | 0.1715 | 69.6558 | 89.0297 | 17.4004 | 6.1297 |
| 0.0059 | 25.0 | 5425 | 0.1710 | 69.6504 | 87.9393 | 17.6101 | 7.2594 |
| 0.0041 | 26.0 | 5642 | 0.1730 | 62.5723 | 87.0114 | 18.2914 | 8.3473 |
| 0.0041 | 27.0 | 5859 | 0.1727 | 65.0893 | 88.5422 | 17.0860 | 6.9456 |
| 0.0031 | 28.0 | 6076 | 0.1725 | 70.9388 | 88.1427 | 16.7191 | 7.3222 |
| 0.0031 | 29.0 | 6293 | 0.1705 | 70.5254 | 88.4779 | 16.8239 | 6.9456 |
| 0.0027 | 30.0 | 6510 | 0.1707 | 71.1488 | 88.5933 | 16.7715 | 6.9038 |
### Framework versions
- Transformers 4.49.0
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0
|