<!-- 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-multiwoz2.1_800-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: 0.3479
- Accuracy: 41.5609
- Num: 7365
- Gen Len: 15.8372
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: 60
- eval_batch_size: 400
- 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 |
---|---|---|---|---|---|---|
1.3224 | 2.22 | 200 | 0.5297 | 27.7246 | 7365 | 15.4379 |
0.581 | 4.44 | 400 | 0.4345 | 33.0547 | 7365 | 15.4058 |
0.498 | 6.67 | 600 | 0.4021 | 34.9997 | 7365 | 15.9775 |
0.4625 | 8.89 | 800 | 0.3876 | 35.7346 | 7365 | 15.8725 |
0.4381 | 11.11 | 1000 | 0.3764 | 36.9032 | 7365 | 15.7621 |
0.4194 | 13.33 | 1200 | 0.3688 | 37.7098 | 7365 | 16.0453 |
0.4061 | 15.56 | 1400 | 0.3642 | 38.3643 | 7365 | 15.5204 |
0.3939 | 17.78 | 1600 | 0.3600 | 38.8761 | 7365 | 15.1166 |
0.3811 | 20.0 | 1800 | 0.3563 | 39.153 | 7365 | 15.5255 |
0.3736 | 22.22 | 2000 | 0.3544 | 39.9885 | 7365 | 15.9825 |
0.3656 | 24.44 | 2200 | 0.3532 | 39.9766 | 7365 | 15.4758 |
0.356 | 26.67 | 2400 | 0.3503 | 40.2686 | 7365 | 15.8242 |
0.3525 | 28.89 | 2600 | 0.3484 | 41.0194 | 7365 | 15.8925 |
0.3463 | 31.11 | 2800 | 0.3484 | 40.7746 | 7365 | 15.8033 |
0.3386 | 33.33 | 3000 | 0.3483 | 40.5603 | 7365 | 15.7708 |
0.3341 | 35.56 | 3200 | 0.3489 | 41.2147 | 7365 | 15.9707 |
0.3291 | 37.78 | 3400 | 0.3485 | 41.4603 | 7365 | 15.8473 |
0.3232 | 40.0 | 3600 | 0.3479 | 41.5609 | 7365 | 15.8372 |
0.3188 | 42.22 | 3800 | 0.3488 | 41.499 | 7365 | 15.8732 |
0.3142 | 44.44 | 4000 | 0.3484 | 41.7224 | 7365 | 15.6745 |
0.3104 | 46.67 | 4200 | 0.3497 | 41.8095 | 7365 | 15.9066 |
0.3079 | 48.89 | 4400 | 0.3495 | 41.8793 | 7365 | 15.7776 |
0.304 | 51.11 | 4600 | 0.3516 | 42.1058 | 7365 | 15.9699 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.5.1
- Tokenizers 0.12.1