<!-- 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-da-multiwoz2.1_400
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: 0.3640
- Accuracy: 41.5248
- Num: 3689
- Gen Len: 15.72
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: 48
- seed: 1799
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Num | Gen Len |
|---|---|---|---|---|---|---|
| 1.182 | 1.16 | 200 | 0.5116 | 29.9369 | 3689 | 13.7774 |
| 0.5435 | 2.33 | 400 | 0.4323 | 33.5836 | 3689 | 15.3494 |
| 0.4654 | 3.49 | 600 | 0.3987 | 35.5517 | 3689 | 15.91 |
| 0.4345 | 4.65 | 800 | 0.3823 | 37.6767 | 3689 | 15.8986 |
| 0.4054 | 5.81 | 1000 | 0.3725 | 38.6302 | 3689 | 15.19 |
| 0.3883 | 6.98 | 1200 | 0.3689 | 38.7642 | 3689 | 15.7896 |
| 0.3694 | 8.14 | 1400 | 0.3685 | 40.0178 | 3689 | 16.3613 |
| 0.3573 | 9.3 | 1600 | 0.3639 | 40.4155 | 3689 | 15.614 |
| 0.3465 | 10.47 | 1800 | 0.3633 | 40.8682 | 3689 | 15.9461 |
| 0.3404 | 11.63 | 2000 | 0.3631 | 40.4788 | 3689 | 16.0949 |
| 0.329 | 12.79 | 2200 | 0.3617 | 41.5163 | 3689 | 15.434 |
| 0.3176 | 13.95 | 2400 | 0.3662 | 41.2127 | 3689 | 15.8075 |
| 0.3129 | 15.12 | 2600 | 0.3665 | 41.2478 | 3689 | 15.5636 |
| 0.3153 | 16.28 | 2800 | 0.3648 | 41.0925 | 3689 | 15.6617 |
| 0.3073 | 17.44 | 3000 | 0.3644 | 41.276 | 3689 | 15.7137 |
| 0.3003 | 18.6 | 3200 | 0.3640 | 41.5248 | 3689 | 15.72 |
| 0.2988 | 19.77 | 3400 | 0.3651 | 41.4459 | 3689 | 15.7322 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.5.1
- Tokenizers 0.12.1
