t5-small-finetuned-text-simplification
This model is a fine-tuned version of t5-small on the wiki_auto_asset_turk dataset. It achieves the following results on the evaluation set:
- Loss: 0.1217
- Rouge2 Precision: 0.5537
- Rouge2 Recall: 0.4251
- Rouge2 Fmeasure: 0.4616
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- 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 | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure |
|---|---|---|---|---|---|---|
| 0.1604 | 1.0 | 15119 | 0.1156 | 0.5567 | 0.4266 | 0.4633 |
| 0.1573 | 2.0 | 30238 | 0.1163 | 0.5534 | 0.4258 | 0.462 |
| 0.1552 | 3.0 | 45357 | 0.1197 | 0.5527 | 0.4244 | 0.4608 |
| 0.1514 | 4.0 | 60476 | 0.1214 | 0.5528 | 0.4257 | 0.4617 |
| 0.1524 | 5.0 | 75595 | 0.1217 | 0.5537 | 0.4251 | 0.4616 |
Framework versions
- Transformers 4.22.0
- Pytorch 1.12.1+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1