<!-- 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-base-da-multiwoz2.0_400-loss-ep100
This model is a fine-tuned version of google/flan-t5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3741
- Accuracy: 39.1797
- Num: 7358
- Gen Len: 15.6147
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: 80
- seed: 1799
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 100
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Num | Gen Len |
---|---|---|---|---|---|---|
1.1208 | 2.33 | 400 | 0.5132 | 26.0596 | 7358 | 14.302 |
0.553 | 4.65 | 800 | 0.4287 | 33.6512 | 7358 | 15.3968 |
0.4783 | 6.98 | 1200 | 0.4007 | 35.3232 | 7358 | 15.8898 |
0.4379 | 9.3 | 1600 | 0.3908 | 36.7949 | 7358 | 15.5749 |
0.4097 | 11.63 | 2000 | 0.3851 | 36.8451 | 7358 | 16.4447 |
0.3859 | 13.95 | 2400 | 0.3770 | 37.9797 | 7358 | 16.2493 |
0.3675 | 16.28 | 2800 | 0.3741 | 39.2162 | 7358 | 16.0883 |
0.3519 | 18.6 | 3200 | 0.3741 | 39.1797 | 7358 | 15.6147 |
0.34 | 20.93 | 3600 | 0.3757 | 40.1516 | 7358 | 15.8101 |
0.3277 | 23.26 | 4000 | 0.3774 | 40.2096 | 7358 | 15.8341 |
0.3181 | 25.58 | 4400 | 0.3755 | 40.3496 | 7358 | 15.4981 |
0.3063 | 27.91 | 4800 | 0.3782 | 40.6828 | 7358 | 15.5501 |
0.2934 | 30.23 | 5200 | 0.3831 | 40.8427 | 7358 | 15.8903 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.5.1
- Tokenizers 0.12.1