<!-- 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-large-SQuAD-qag-ep6
This model is a fine-tuned version of t5-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8950
- Rouge1: 41.0251
- Rouge2: 19.2729
- Rougel: 37.4514
- Rougelsum: 37.4839
- F1: 20.0159
- Exact Match: 14.2719
- Gen Len: 18.418
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: 24
- eval_batch_size: 48
- seed: 1799
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | F1 | Exact Match | Gen Len |
|---|---|---|---|---|---|---|---|---|---|---|
| 1.1752 | 0.51 | 400 | 0.9542 | 39.054 | 17.5678 | 35.471 | 35.4744 | 16.7879 | 10.8853 | 18.5186 |
| 1.0232 | 1.02 | 800 | 0.9232 | 40.062 | 18.493 | 36.4108 | 36.4335 | 18.7316 | 12.9173 | 18.4543 |
| 0.9641 | 1.52 | 1200 | 0.9112 | 40.6835 | 19.0484 | 37.0867 | 37.1329 | 19.3326 | 14.03 | 18.4078 |
| 0.9422 | 2.03 | 1600 | 0.9044 | 40.9087 | 19.2513 | 37.3069 | 37.3196 | 19.999 | 14.2719 | 18.4383 |
| 0.8989 | 2.54 | 2000 | 0.9026 | 41.0666 | 19.4921 | 37.4916 | 37.5115 | 21.0451 | 14.8041 | 18.3469 |
| 0.9015 | 3.05 | 2400 | 0.8986 | 40.471 | 18.8468 | 36.9384 | 36.9357 | 19.3467 | 13.8365 | 18.4557 |
| 0.8679 | 3.56 | 2800 | 0.8950 | 41.0251 | 19.2729 | 37.4514 | 37.4839 | 20.0159 | 14.2719 | 18.418 |
| 0.8521 | 4.07 | 3200 | 0.8982 | 41.1633 | 19.6433 | 37.6253 | 37.6541 | 21.0075 | 15.046 | 18.3807 |
| 0.8362 | 4.57 | 3600 | 0.8972 | 40.9127 | 19.4618 | 37.4 | 37.4431 | 20.9531 | 14.8524 | 18.3503 |
| 0.8415 | 5.08 | 4000 | 0.8970 | 41.2265 | 19.7241 | 37.7405 | 37.7653 | 21.5704 | 15.1911 | 18.388 |
| 0.8312 | 5.59 | 4400 | 0.8966 | 40.9824 | 19.4891 | 37.533 | 37.5485 | 20.8781 | 14.7073 | 18.3967 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.11.0+cu113
- Datasets 2.5.1
- Tokenizers 0.12.1
