distilbert-uncased-finetuned-toxic-comments-detection
This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1520
- Accuracy: 0.95
- Precision: 0.7391
- Recall: 0.8095
- F1: 0.7727
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: 16
- eval_batch_size: 8
- 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 |
|---|---|---|---|---|---|---|---|
| 0.3892 | 1.0 | 50 | 0.2808 | 0.895 | 0.0 | 0.0 | 0.0 |
| 0.219 | 2.0 | 100 | 0.1732 | 0.93 | 0.8182 | 0.4286 | 0.5625 |
| 0.1313 | 3.0 | 150 | 0.1515 | 0.95 | 0.7391 | 0.8095 | 0.7727 |
| 0.0924 | 4.0 | 200 | 0.1520 | 0.95 | 0.7391 | 0.8095 | 0.7727 |
| 0.0749 | 5.0 | 250 | 0.1540 | 0.96 | 0.8095 | 0.8095 | 0.8095 |
Framework versions
- Transformers 4.30.1
- Pytorch 2.0.1+cu118
- Tokenizers 0.13.3