<!-- 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. -->
distilbert-base-uncasedfinetuned-clinc
This model is a fine-tuned version of distilbert-base-uncased on the clinc_oos dataset. It achieves the following results on the evaluation set:
- Loss: 0.2855
- Accuracy: 0.9474
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: 48
- eval_batch_size: 48
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
4.1322 | 1.0 | 318 | 3.0871 | 0.7513 |
2.349 | 2.0 | 636 | 1.5284 | 0.8571 |
1.1416 | 3.0 | 954 | 0.7639 | 0.9165 |
0.5644 | 4.0 | 1272 | 0.4679 | 0.9345 |
0.3118 | 5.0 | 1590 | 0.3526 | 0.9445 |
0.2018 | 6.0 | 1908 | 0.3124 | 0.9445 |
0.149 | 7.0 | 2226 | 0.2924 | 0.9481 |
0.1239 | 8.0 | 2544 | 0.2889 | 0.9484 |
0.1117 | 9.0 | 2862 | 0.2880 | 0.9471 |
0.1073 | 10.0 | 3180 | 0.2855 | 0.9474 |
Framework versions
- Transformers 4.16.2
- Pytorch 2.0.1+cu118
- Datasets 1.16.1
- Tokenizers 0.14.0