xlm-roberta-base-amh-finetuned-augmentation-LUNAR-finetuned-augmentation
This model is a fine-tuned version of sercetexam9/xlm-roberta-base-amh-finetuned-augmentation-LUNAR on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1407
- F1: 0.8442
- Roc Auc: 0.9039
- Accuracy: 0.7935
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
|---|---|---|---|---|---|---|
| 0.1268 | 1.0 | 208 | 0.1278 | 0.8404 | 0.9046 | 0.8213 |
| 0.1243 | 2.0 | 416 | 0.1407 | 0.8442 | 0.9039 | 0.7935 |
| 0.092 | 3.0 | 624 | 0.1405 | 0.8402 | 0.8974 | 0.8007 |
| 0.0826 | 4.0 | 832 | 0.1726 | 0.8193 | 0.8892 | 0.7597 |
| 0.0663 | 5.0 | 1040 | 0.1503 | 0.8239 | 0.8822 | 0.7874 |
| 0.0536 | 6.0 | 1248 | 0.1568 | 0.8200 | 0.8973 | 0.7995 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0