metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: normal_en_to_poe_translator
results: []
normal_en_to_poe_translator
This model is a fine-tuned version of t5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0203
- Bleu: 17.4122
- Gen Len: 16.9451
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 |
|---|---|---|---|---|---|
| 1.3708 | 1.0 | 1182 | 1.2112 | 16.0712 | 16.9251 |
| 1.2776 | 2.0 | 2364 | 1.1458 | 16.5284 | 16.9233 |
| 1.2014 | 3.0 | 3546 | 1.1060 | 16.7928 | 16.9259 |
| 1.1637 | 4.0 | 4728 | 1.0783 | 16.9978 | 16.9317 |
| 1.1436 | 5.0 | 5910 | 1.0587 | 17.131 | 16.936 |
| 1.109 | 6.0 | 7092 | 1.0427 | 17.2265 | 16.9386 |
| 1.0912 | 7.0 | 8274 | 1.0324 | 17.2943 | 16.938 |
| 1.0751 | 8.0 | 9456 | 1.0257 | 17.3605 | 16.9432 |
| 1.0642 | 9.0 | 10638 | 1.0216 | 17.395 | 16.9426 |
| 1.0567 | 10.0 | 11820 | 1.0203 | 17.4122 | 16.9451 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3