PRE-xlnet-large-cased-finetuned-augmentation-2
This model is a fine-tuned version of xlnet-large-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3383
- F1: 0.7385
- Roc Auc: 0.8538
- Accuracy: 0.7857
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 |
|---|---|---|---|---|---|---|
| 0.324 | 1.0 | 389 | 0.2935 | 0.1551 | 0.5725 | 0.5663 |
| 0.2498 | 2.0 | 778 | 0.1995 | 0.5340 | 0.7415 | 0.7072 |
| 0.2112 | 3.0 | 1167 | 0.1986 | 0.5552 | 0.7607 | 0.6950 |
| 0.2189 | 4.0 | 1556 | 0.1733 | 0.6698 | 0.8068 | 0.7426 |
| 0.1263 | 5.0 | 1945 | 0.1857 | 0.6895 | 0.8174 | 0.7516 |
| 0.1076 | 6.0 | 2334 | 0.1919 | 0.6977 | 0.8059 | 0.7671 |
| 0.0683 | 7.0 | 2723 | 0.2189 | 0.7003 | 0.8168 | 0.7625 |
| 0.0405 | 8.0 | 3112 | 0.2597 | 0.7169 | 0.8385 | 0.7671 |
| 0.0212 | 9.0 | 3501 | 0.2865 | 0.7172 | 0.8446 | 0.7606 |
| 0.0349 | 10.0 | 3890 | 0.3011 | 0.6998 | 0.8268 | 0.7638 |
| 0.0198 | 11.0 | 4279 | 0.3188 | 0.7204 | 0.8393 | 0.7754 |
| 0.0222 | 12.0 | 4668 | 0.3385 | 0.7347 | 0.8589 | 0.7780 |
| 0.0037 | 13.0 | 5057 | 0.3355 | 0.7330 | 0.8467 | 0.7851 |
| 0.0071 | 14.0 | 5446 | 0.3383 | 0.7385 | 0.8538 | 0.7857 |
| 0.0011 | 15.0 | 5835 | 0.3536 | 0.7301 | 0.8430 | 0.7831 |
| 0.0045 | 16.0 | 6224 | 0.3494 | 0.7358 | 0.8453 | 0.7857 |
| 0.0013 | 17.0 | 6613 | 0.3555 | 0.7339 | 0.8473 | 0.7844 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0