CodeDetect-RoBERTa-T1
This model is a fine-tuned version of FacebookAI/roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0912
- Accuracy: 0.9745
- F1: 0.9745
- Precision: 0.9746
- Recall: 0.9744
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 0.1
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.0978 | 1.0 | 16897 | 0.0952 | 0.9660 | 0.9659 | 0.9659 | 0.9659 |
| 0.0779 | 2.0 | 33794 | 0.0811 | 0.9705 | 0.9705 | 0.9707 | 0.9703 |
| 0.0629 | 3.0 | 50691 | 0.0837 | 0.9721 | 0.9721 | 0.9724 | 0.9718 |
| 0.0529 | 4.0 | 67588 | 0.0796 | 0.9750 | 0.9749 | 0.9749 | 0.9750 |
| 0.0397 | 5.0 | 84485 | 0.0912 | 0.9745 | 0.9745 | 0.9746 | 0.9744 |
Framework versions
- Transformers 5.13.1
- Pytorch 2.11.0+cu130
- Datasets 4.0.0
- Tokenizers 0.22.2