<!-- 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-extraction-all-dm_2000-ep25-nonstop
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: 1.6447
- Hint Hit Num: 2.581
- Hint Precision: 0.491
- Num: 5.255
- Gen Len: 18.407
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: 96
- seed: 1799
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 25
Training results
| Training Loss | Epoch | Step | Validation Loss | Hint Hit Num | Hint Precision | Num | Gen Len |
|---|---|---|---|---|---|---|---|
| 2.2187 | 2.38 | 200 | 1.5459 | 2.651 | 0.4887 | 5.437 | 18.784 |
| 1.8908 | 4.76 | 400 | 1.5341 | 2.637 | 0.4868 | 5.413 | 18.721 |
| 1.7562 | 7.14 | 600 | 1.5352 | 2.569 | 0.4746 | 5.406 | 18.684 |
| 1.6683 | 9.52 | 800 | 1.5501 | 2.609 | 0.4811 | 5.414 | 18.658 |
| 1.5782 | 11.9 | 1000 | 1.5733 | 2.58 | 0.4852 | 5.306 | 18.536 |
| 1.517 | 14.29 | 1200 | 1.5896 | 2.619 | 0.491 | 5.32 | 18.579 |
| 1.464 | 16.67 | 1400 | 1.6124 | 2.632 | 0.4992 | 5.283 | 18.414 |
| 1.4189 | 19.05 | 1600 | 1.6326 | 2.548 | 0.4863 | 5.24 | 18.436 |
| 1.3901 | 21.43 | 1800 | 1.6408 | 2.563 | 0.4881 | 5.247 | 18.408 |
| 1.3845 | 23.81 | 2000 | 1.6429 | 2.601 | 0.4931 | 5.266 | 18.424 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.5.1
- Tokenizers 0.12.1
