proof_eval1
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3242
- Accuracy: 0.8794
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 392 | 0.3181 | 0.8716 |
| 0.4191 | 2.0 | 784 | 0.3347 | 0.8671 |
| 0.2834 | 3.0 | 1176 | 0.2852 | 0.8788 |
| 0.2332 | 4.0 | 1568 | 0.3108 | 0.8844 |
| 0.2332 | 5.0 | 1960 | 0.3242 | 0.8794 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2