climate_un2
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: 0.0929
- Accuracy: 0.9860
- F1 Macro: 0.9804
- Accuracy Balanced: 0.9856
- F1 Micro: 0.9860
- Precision Macro: 0.9755
- Recall Macro: 0.9856
- Precision Micro: 0.9860
- Recall Micro: 0.9860
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: 80
- 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
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Accuracy Balanced | F1 Micro | Precision Macro | Recall Macro | Precision Micro | Recall Micro |
---|---|---|---|---|---|---|---|---|---|---|---|
No log | 1.0 | 167 | 0.0673 | 0.9851 | 0.9790 | 0.9796 | 0.9851 | 0.9783 | 0.9796 | 0.9851 | 0.9851 |
No log | 2.0 | 334 | 0.0732 | 0.9834 | 0.9768 | 0.9839 | 0.9834 | 0.9702 | 0.9839 | 0.9834 | 0.9834 |
0.1178 | 3.0 | 501 | 0.0617 | 0.9860 | 0.9804 | 0.9856 | 0.9860 | 0.9755 | 0.9856 | 0.9860 | 0.9860 |
0.1178 | 4.0 | 668 | 0.0593 | 0.9886 | 0.9840 | 0.9873 | 0.9886 | 0.9809 | 0.9873 | 0.9886 | 0.9886 |
0.1178 | 5.0 | 835 | 0.0822 | 0.9843 | 0.9780 | 0.9844 | 0.9843 | 0.9720 | 0.9844 | 0.9843 | 0.9843 |
0.0088 | 6.0 | 1002 | 0.0810 | 0.9878 | 0.9829 | 0.9880 | 0.9878 | 0.9779 | 0.9880 | 0.9878 | 0.9878 |
0.0088 | 7.0 | 1169 | 0.0904 | 0.9860 | 0.9804 | 0.9856 | 0.9860 | 0.9755 | 0.9856 | 0.9860 | 0.9860 |
0.0088 | 8.0 | 1336 | 0.0927 | 0.9860 | 0.9804 | 0.9856 | 0.9860 | 0.9755 | 0.9856 | 0.9860 | 0.9860 |
0.0006 | 9.0 | 1503 | 0.0936 | 0.9860 | 0.9805 | 0.9869 | 0.9860 | 0.9744 | 0.9869 | 0.9860 | 0.9860 |
0.0006 | 10.0 | 1670 | 0.0929 | 0.9860 | 0.9804 | 0.9856 | 0.9860 | 0.9755 | 0.9856 | 0.9860 | 0.9860 |
Framework versions
- Transformers 4.35.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.7
- Tokenizers 0.14.1
- Downloads last month
- 2