<!-- 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-multiwoz2.0_80-loss-ep50
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: 0.4604
- Accuracy: 33.2618
- Num: 7358
- Gen Len: 15.5321
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: 192
- seed: 1799
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Num | Gen Len |
|---|---|---|---|---|---|---|
| 1.1359 | 8.0 | 200 | 0.4929 | 27.6197 | 7358 | 14.857 |
| 0.474 | 16.0 | 400 | 0.4604 | 33.2618 | 7358 | 15.5321 |
| 0.3764 | 24.0 | 600 | 0.4672 | 34.9136 | 7358 | 15.5559 |
| 0.323 | 32.0 | 800 | 0.4783 | 35.5058 | 7358 | 15.7635 |
| 0.2858 | 40.0 | 1000 | 0.5008 | 35.7085 | 7358 | 15.9337 |
| 0.2709 | 48.0 | 1200 | 0.5069 | 35.8528 | 7358 | 15.981 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.5.1
- Tokenizers 0.12.1
