<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
covid-general-news-bert
This model is a fine-tuned version of bvrau/covid-twitter-bert-v2-struth on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0688
- Accuracy: 0.9774
- Precision: 0.9781
- Recall: 0.9738
- F1: 0.9760
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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.2183 | 1.0 | 365 | 0.0688 | 0.9774 | 0.9781 | 0.9738 | 0.9760 |
| 0.0783 | 2.0 | 730 | 0.0754 | 0.9842 | 0.9812 | 0.9855 | 0.9833 |
| 0.0354 | 3.0 | 1095 | 0.0766 | 0.9856 | 0.9785 | 0.9913 | 0.9848 |
| 0.0185 | 4.0 | 1460 | 0.0956 | 0.9822 | 0.9715 | 0.9913 | 0.9813 |
| 0.0227 | 5.0 | 1825 | 0.0693 | 0.9870 | 0.9827 | 0.9898 | 0.9862 |
| 0.0084 | 6.0 | 2190 | 0.0870 | 0.9849 | 0.9926 | 0.9753 | 0.9839 |
| 0.0021 | 7.0 | 2555 | 0.0729 | 0.9877 | 0.9883 | 0.9855 | 0.9869 |
| 0.0002 | 8.0 | 2920 | 0.1197 | 0.9808 | 0.9688 | 0.9913 | 0.9799 |
| 0.0033 | 9.0 | 3285 | 0.0768 | 0.9884 | 0.9912 | 0.9840 | 0.9876 |
| 0.0009 | 10.0 | 3650 | 0.1013 | 0.9863 | 0.9869 | 0.9840 | 0.9854 |
| 0.0 | 11.0 | 4015 | 0.1069 | 0.9863 | 0.9869 | 0.9840 | 0.9854 |
| 0.0 | 12.0 | 4380 | 0.1124 | 0.9856 | 0.9854 | 0.9840 | 0.9847 |
| 0.0 | 13.0 | 4745 | 0.1175 | 0.9849 | 0.9854 | 0.9826 | 0.9840 |
| 0.0 | 14.0 | 5110 | 0.1221 | 0.9849 | 0.9854 | 0.9826 | 0.9840 |
| 0.0 | 15.0 | 5475 | 0.1256 | 0.9849 | 0.9854 | 0.9826 | 0.9840 |
| 0.0 | 16.0 | 5840 | 0.1286 | 0.9849 | 0.9854 | 0.9826 | 0.9840 |
| 0.0 | 17.0 | 6205 | 0.1300 | 0.9856 | 0.9854 | 0.9840 | 0.9847 |
| 0.0 | 18.0 | 6570 | 0.1293 | 0.9849 | 0.9854 | 0.9826 | 0.9840 |
| 0.0 | 19.0 | 6935 | 0.1304 | 0.9849 | 0.9854 | 0.9826 | 0.9840 |
| 0.0 | 20.0 | 7300 | 0.1308 | 0.9849 | 0.9854 | 0.9826 | 0.9840 |
Framework versions
- Transformers 4.22.2
- Pytorch 1.12.1+cu113
- Datasets 2.5.1
- Tokenizers 0.12.1
