<!-- 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_800
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: 0.3334
- Accuracy: 44.0942
- Num: 3689
- Gen Len: 15.7842
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 |
---|---|---|---|---|---|---|
1.1927 | 0.6 | 200 | 0.5022 | 31.6941 | 3689 | 15.2413 |
0.5504 | 1.19 | 400 | 0.4375 | 33.8508 | 3689 | 16.6571 |
0.4773 | 1.79 | 600 | 0.3986 | 34.0235 | 3689 | 14.4118 |
0.4458 | 2.38 | 800 | 0.3830 | 35.6435 | 3689 | 14.3944 |
0.4241 | 2.98 | 1000 | 0.3654 | 38.1415 | 3689 | 15.0249 |
0.4038 | 3.57 | 1200 | 0.3551 | 38.6647 | 3689 | 16.1971 |
0.3824 | 4.17 | 1400 | 0.3480 | 39.7139 | 3689 | 15.4738 |
0.3781 | 4.76 | 1600 | 0.3512 | 40.5284 | 3689 | 15.4681 |
0.3637 | 5.36 | 1800 | 0.3470 | 41.012 | 3689 | 15.6362 |
0.3649 | 5.95 | 2000 | 0.3403 | 40.543 | 3689 | 15.7314 |
0.3498 | 6.55 | 2200 | 0.3432 | 42.0574 | 3689 | 15.9043 |
0.3442 | 7.14 | 2400 | 0.3407 | 41.7062 | 3689 | 15.2239 |
0.3381 | 7.74 | 2600 | 0.3386 | 41.9579 | 3689 | 15.3402 |
0.3322 | 8.33 | 2800 | 0.3358 | 42.7824 | 3689 | 15.7652 |
0.33 | 8.93 | 3000 | 0.3336 | 42.4657 | 3689 | 15.4942 |
0.3238 | 9.52 | 3200 | 0.3347 | 42.8539 | 3689 | 15.7726 |
0.3162 | 10.12 | 3400 | 0.3371 | 42.9605 | 3689 | 15.3914 |
0.3168 | 10.71 | 3600 | 0.3351 | 42.7597 | 3689 | 15.5072 |
0.3154 | 11.31 | 3800 | 0.3334 | 44.0942 | 3689 | 15.7842 |
0.3032 | 11.9 | 4000 | 0.3375 | 43.5387 | 3689 | 15.656 |
0.3032 | 12.5 | 4200 | 0.3363 | 43.7886 | 3689 | 15.3608 |
0.3003 | 13.1 | 4400 | 0.3393 | 43.4861 | 3689 | 15.82 |
0.2963 | 13.69 | 4600 | 0.3367 | 43.7505 | 3689 | 15.3513 |
0.2961 | 14.29 | 4800 | 0.3350 | 43.7886 | 3689 | 15.5232 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.5.1
- Tokenizers 0.12.1