CodeGenDetect-Roberta_Lora
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: 0.0373
- Accuracy: 0.9884
- F1: 0.9884
- Precision: 0.9884
- Recall: 0.9884
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: 128
- eval_batch_size: 128
- 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: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.0526 | 1.02 | 4000 | 0.0545 | 0.9825 | 0.9825 | 0.9825 | 0.9825 |
| 0.0402 | 2.05 | 8000 | 0.0401 | 0.9872 | 0.9872 | 0.9872 | 0.9872 |
| 0.0349 | 3.07 | 12000 | 0.0416 | 0.9870 | 0.9871 | 0.9871 | 0.9870 |
| 0.0404 | 4.1 | 16000 | 0.0373 | 0.9884 | 0.9884 | 0.9884 | 0.9884 |
Framework versions
- PEFT 0.9.0
- Transformers 4.38.2
- Pytorch 2.5.1+rocm6.2
- Datasets 2.21.0
- Tokenizers 0.15.2