distilbert-base-uncased-finetuned-dwnews-categories
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8331
- F1: 0.7310
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: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 2.0432 | 0.3 | 30 | 1.8112 | 0.2093 |
| 1.7228 | 0.6 | 60 | 1.4949 | 0.3650 |
| 1.3799 | 0.9 | 90 | 1.2691 | 0.5838 |
| 1.2261 | 1.2 | 120 | 1.1287 | 0.6345 |
| 1.0695 | 1.5 | 150 | 1.0383 | 0.6723 |
| 0.9634 | 1.8 | 180 | 0.9570 | 0.7279 |
| 0.9289 | 2.1 | 210 | 0.9106 | 0.7435 |
| 0.8258 | 2.4 | 240 | 0.9380 | 0.7130 |
| 0.7692 | 2.7 | 270 | 0.8708 | 0.7262 |
| 0.7542 | 3.0 | 300 | 0.8568 | 0.7350 |
| 0.6584 | 3.3 | 330 | 0.8447 | 0.7368 |
| 0.5871 | 3.6 | 360 | 0.8517 | 0.7226 |
| 0.6528 | 3.9 | 390 | 0.8471 | 0.7290 |
| 0.5805 | 4.2 | 420 | 0.8085 | 0.7291 |
| 0.5904 | 4.5 | 450 | 0.8331 | 0.7310 |
| 0.4877 | 4.8 | 480 | 0.8334 | 0.7209 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.10.0
- Tokenizers 0.13.2