<!-- 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. -->
userutterance_classification_verplus
This model is a fine-tuned version of microsoft/deberta-v3-base on the clinc_oos dataset. It achieves the following results on the evaluation set:
- Loss: 0.2270
- Accuracy: 0.9619
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: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 5.0219 | 0.21 | 200 | 4.9813 | 0.0077 |
| 4.8915 | 0.42 | 400 | 4.5741 | 0.1155 |
| 4.2736 | 0.63 | 600 | 3.5359 | 0.4719 |
| 3.2701 | 0.84 | 800 | 2.4291 | 0.7429 |
| 2.3578 | 1.05 | 1000 | 1.5793 | 0.8413 |
| 1.5695 | 1.26 | 1200 | 1.0029 | 0.8994 |
| 1.0412 | 1.47 | 1400 | 0.6475 | 0.9187 |
| 0.7034 | 1.68 | 1600 | 0.4439 | 0.9303 |
| 0.501 | 1.89 | 1800 | 0.3400 | 0.9381 |
| 0.3187 | 2.1 | 2000 | 0.2793 | 0.9439 |
| 0.2185 | 2.31 | 2200 | 0.2538 | 0.9490 |
| 0.1669 | 2.52 | 2400 | 0.2210 | 0.9523 |
| 0.1081 | 2.73 | 2600 | 0.2225 | 0.9519 |
| 0.1004 | 2.94 | 2800 | 0.2136 | 0.9555 |
| 0.0665 | 3.14 | 3000 | 0.2078 | 0.9561 |
| 0.0509 | 3.35 | 3200 | 0.2155 | 0.9568 |
| 0.05 | 3.56 | 3400 | 0.2107 | 0.9581 |
| 0.0527 | 3.77 | 3600 | 0.2171 | 0.9568 |
| 0.0447 | 3.98 | 3800 | 0.2128 | 0.9590 |
| 0.0259 | 4.19 | 4000 | 0.2099 | 0.9587 |
| 0.0279 | 4.4 | 4200 | 0.2179 | 0.9577 |
| 0.0176 | 4.61 | 4400 | 0.2191 | 0.9574 |
| 0.0288 | 4.82 | 4600 | 0.2216 | 0.9590 |
| 0.0328 | 5.03 | 4800 | 0.2237 | 0.9606 |
| 0.0154 | 5.24 | 5000 | 0.2241 | 0.9616 |
| 0.0157 | 5.45 | 5200 | 0.2265 | 0.9603 |
| 0.023 | 5.66 | 5400 | 0.2276 | 0.9613 |
| 0.0178 | 5.87 | 5600 | 0.2270 | 0.9619 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
