edos-2023-baseline-distilbert-base-uncased-label_sexist
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: 0.4852
- F1: 0.7874
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
- lr_scheduler_warmup_steps: 5
- num_epochs: 8
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 0.4199 | 1.14 | 400 | 0.3911 | 0.7571 |
| 0.293 | 2.29 | 800 | 0.3778 | 0.7899 |
| 0.2348 | 3.43 | 1200 | 0.4102 | 0.7894 |
| 0.1895 | 4.57 | 1600 | 0.4417 | 0.7835 |
| 0.1392 | 5.71 | 2000 | 0.4852 | 0.7874 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2