t5-mt-en-ca-new
This model is a fine-tuned version of t5-small on the opus_books dataset. It achieves the following results on the evaluation set:
- Loss: 3.4508
- Bleu: 0.2189
- Gen Len: 17.5396
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 | 231 | 3.6893 | 0.0917 | 17.7687 |
| No log | 2.0 | 462 | 3.5711 | 0.1244 | 17.6591 |
| 3.9065 | 3.0 | 693 | 3.5008 | 0.1889 | 17.6059 |
| 3.9065 | 4.0 | 924 | 3.4631 | 0.2197 | 17.5244 |
| 3.7288 | 5.0 | 1155 | 3.4508 | 0.2189 | 17.5396 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3