<!-- 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_80-new
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: 0.5840
- Accuracy: 36.4554
- Num: 3689
- Gen Len: 15.7091
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: 8
- eval_batch_size: 24
- 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 |
---|---|---|---|---|---|---|
1.353 | 2.74 | 200 | 0.5557 | 28.2706 | 3689 | 16.434 |
0.578 | 5.48 | 400 | 0.4761 | 32.3269 | 3689 | 16.3172 |
0.463 | 8.22 | 600 | 0.4581 | 34.1789 | 3689 | 16.6969 |
0.402 | 10.96 | 800 | 0.4498 | 34.5196 | 3689 | 15.9797 |
0.3527 | 13.7 | 1000 | 0.4735 | 33.9929 | 3689 | 16.2041 |
0.3087 | 16.44 | 1200 | 0.5051 | 35.8301 | 3689 | 16.1225 |
0.2695 | 19.18 | 1400 | 0.5304 | 35.6991 | 3689 | 16.0713 |
0.2448 | 21.92 | 1600 | 0.5390 | 35.9178 | 3689 | 16.17 |
0.2101 | 24.66 | 1800 | 0.5840 | 36.4554 | 3689 | 15.7091 |
0.1803 | 27.4 | 2000 | 0.6295 | 35.8091 | 3689 | 15.7327 |
0.1683 | 30.14 | 2200 | 0.6311 | 35.8789 | 3689 | 15.5169 |
0.1497 | 32.88 | 2400 | 0.6851 | 35.8932 | 3689 | 15.4825 |
0.1285 | 35.62 | 2600 | 0.7251 | 35.4655 | 3689 | 15.2909 |
0.1179 | 38.36 | 2800 | 0.7664 | 35.8041 | 3689 | 15.3185 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.5.1
- Tokenizers 0.12.1