my_model
This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2877
- Bleu: 0.0001
- Gen Len: 17.7878
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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 0.3656 | 1.0 | 1000 | 0.3225 | 0.0 | 18.175 |
| 0.3395 | 2.0 | 2000 | 0.3070 | 0.0 | 18.193 |
| 0.3266 | 3.0 | 3000 | 0.2998 | 0.0 | 18.0902 |
| 0.3142 | 4.0 | 4000 | 0.2957 | 0.0 | 17.8615 |
| 0.3155 | 5.0 | 5000 | 0.2925 | 0.0 | 17.8932 |
| 0.3045 | 6.0 | 6000 | 0.2908 | 0.0 | 17.8382 |
| 0.3093 | 7.0 | 7000 | 0.2894 | 0.0 | 17.8515 |
| 0.3012 | 8.0 | 8000 | 0.2885 | 0.0001 | 17.827 |
| 0.3007 | 9.0 | 9000 | 0.2878 | 0.0 | 17.7822 |
| 0.302 | 10.0 | 10000 | 0.2877 | 0.0001 | 17.7878 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3