<!-- 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. -->
flan-t5-large-SQuAD-qag-ep6
This model is a fine-tuned version of google/flan-t5-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8670
- Rouge1: 40.8926
- Rouge2: 19.5196
- Rougel: 37.4672
- Rougelsum: 37.4713
- F1: 21.4247
- Exact Match: 15.1427
- Gen Len: 18.3706
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 |
|---|---|---|---|---|---|---|---|---|---|---|
| 1.0439 | 0.51 | 400 | 0.8849 | 39.8979 | 18.5242 | 36.3173 | 36.3341 | 20.5015 | 13.8365 | 18.4069 |
| 0.9761 | 1.02 | 800 | 0.8748 | 40.0168 | 18.6616 | 36.6103 | 36.6343 | 19.4625 | 13.9332 | 18.4359 |
| 0.9358 | 1.52 | 1200 | 0.8709 | 40.6382 | 19.6212 | 37.2797 | 37.3127 | 20.2207 | 14.2235 | 18.3957 |
| 0.919 | 2.03 | 1600 | 0.8697 | 41.2054 | 20.0197 | 37.7826 | 37.8021 | 21.7337 | 15.4814 | 18.3633 |
| 0.8849 | 2.54 | 2000 | 0.8695 | 41.3385 | 19.9503 | 37.9501 | 37.9707 | 22.0787 | 15.6265 | 18.3507 |
| 0.8869 | 3.05 | 2400 | 0.8674 | 41.2089 | 19.7348 | 37.6962 | 37.7184 | 21.3812 | 14.9492 | 18.4122 |
| 0.8617 | 3.56 | 2800 | 0.8670 | 40.8926 | 19.5196 | 37.4672 | 37.4713 | 21.4247 | 15.1427 | 18.3706 |
| 0.8486 | 4.07 | 3200 | 0.8678 | 40.856 | 19.6152 | 37.4585 | 37.4462 | 22.0102 | 15.5781 | 18.3522 |
| 0.8367 | 4.57 | 3600 | 0.8678 | 40.9533 | 19.7798 | 37.5764 | 37.5818 | 21.769 | 15.7233 | 18.328 |
| 0.8442 | 5.08 | 4000 | 0.8671 | 41.2103 | 19.987 | 37.8326 | 37.8498 | 22.3873 | 16.1587 | 18.3314 |
| 0.8317 | 5.59 | 4400 | 0.8673 | 41.0444 | 19.7886 | 37.6754 | 37.6706 | 22.0032 | 15.7716 | 18.3208 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.11.0+cu113
- Datasets 2.5.1
- Tokenizers 0.12.1
