agnews-distilbert
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: 0.2961
- Accuracy: 0.924
- Precision: 0.9244
- Recall: 0.924
- F1: 0.9241
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: 8
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.216 | 1.0 | 1250 | 0.2758 | 0.91 | 0.9114 | 0.91 | 0.9099 |
| 0.1217 | 2.0 | 2500 | 0.2805 | 0.9195 | 0.9207 | 0.9195 | 0.9197 |
| 0.1281 | 3.0 | 3750 | 0.2961 | 0.924 | 0.9244 | 0.924 | 0.9241 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.11.0+cu128
- Datasets 2.20.0
- Tokenizers 0.19.1