<!-- 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. -->
my_awesome_sindarin_model
This model is a fine-tuned version of t5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.1459
- Bleu: 0.3288
- Gen Len: 12.348
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| No log | 1.0 | 199 | 3.6560 | 0.0612 | 17.4414 |
| No log | 2.0 | 398 | 3.4073 | 0.1168 | 12.5788 |
| 3.9239 | 3.0 | 597 | 3.2573 | 0.2427 | 12.3178 |
| 3.9239 | 4.0 | 796 | 3.1767 | 0.2812 | 12.2585 |
| 3.9239 | 5.0 | 995 | 3.1459 | 0.3288 | 12.348 |
Framework versions
- Transformers 4.28.1
- Pytorch 1.13.1+cu116
- Datasets 2.11.0
- Tokenizers 0.13.3
