<!-- 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. -->
t5-small-finetuned-la-to-en
This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.2297
- Bleu: 5.8915
- Gen Len: 16.2252
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 |
---|---|---|---|---|---|
3.0883 | 1.0 | 4384 | 2.7499 | 2.8172 | 16.4068 |
2.8854 | 2.0 | 8768 | 2.5664 | 3.8141 | 16.4581 |
2.746 | 3.0 | 13152 | 2.4524 | 4.3903 | 16.3977 |
2.6617 | 4.0 | 17536 | 2.3761 | 4.7858 | 16.3473 |
2.6185 | 5.0 | 21920 | 2.3205 | 5.2502 | 16.3161 |
2.573 | 6.0 | 26304 | 2.2763 | 5.4374 | 16.2916 |
2.5285 | 7.0 | 30688 | 2.2489 | 5.628 | 16.2875 |
2.4944 | 8.0 | 35072 | 2.2276 | 5.7201 | 16.291 |
2.4749 | 9.0 | 39456 | 2.2164 | 5.8387 | 16.2795 |
2.4741 | 10.0 | 43840 | 2.2129 | 5.8654 | 16.2789 |
Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1