transformer_model
This model is a fine-tuned version of t5-small on the aslg_pc12 dataset. It achieves the following results on the evaluation set:
- Loss: 0.5811
- Bleu: 56.4281
- Gen Len: 15.5526
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: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 1.094 | 1.0 | 4386 | 0.7098 | 50.634 | 15.6095 |
| 0.9099 | 2.0 | 8772 | 0.6072 | 55.2182 | 15.5592 |
| 0.8697 | 3.0 | 13158 | 0.5811 | 56.4281 | 15.5526 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2