proof_eval2-distilbert
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3134
- Accuracy: 0.8800
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 | 0.3296 | 0.8693 |
| 0.4457 | 2.0 | 784 | 0.3070 | 0.8755 |
| 0.2966 | 3.0 | 1176 | 0.2910 | 0.8788 |
| 0.244 | 4.0 | 1568 | 0.3055 | 0.8844 |
| 0.244 | 5.0 | 1960 | 0.3134 | 0.8800 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2