distilbert-profane-final
This model is a fine-tuned version of distilbert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2773
- Accuracy: 0.8992
- Precision: 0.8261
- Recall: 0.7987
- F1: 0.8114
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: 1e-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
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 296 | 0.2862 | 0.8907 | 0.8230 | 0.7528 | 0.7807 |
| 0.3379 | 2.0 | 592 | 0.2650 | 0.9097 | 0.8748 | 0.7778 | 0.8148 |
| 0.3379 | 3.0 | 888 | 0.2632 | 0.9049 | 0.8417 | 0.7999 | 0.8185 |
| 0.221 | 4.0 | 1184 | 0.2772 | 0.8916 | 0.8055 | 0.8055 | 0.8055 |
| 0.221 | 5.0 | 1480 | 0.2773 | 0.8992 | 0.8261 | 0.7987 | 0.8114 |
Framework versions
- Transformers 4.24.0.dev0
- Pytorch 1.11.0+cu102
- Datasets 2.6.1
- Tokenizers 0.13.1