<!-- 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-augment-tweet-bert-large-e8-noweight
This model is a fine-tuned version of digitalepidemiologylab/covid-twitter-bert-v2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2396
- Accuracy: 0.9714
- F1: 0.9249
- Precision: 0.9095
- Recall: 0.9409
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: 8
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
---|---|---|---|---|---|---|---|
No log | 1.0 | 408 | 0.1663 | 0.9419 | 0.8609 | 0.78 | 0.9606 |
0.2202 | 2.0 | 816 | 0.1532 | 0.9594 | 0.8957 | 0.8630 | 0.9310 |
0.0794 | 3.0 | 1224 | 0.1745 | 0.9687 | 0.9167 | 0.9122 | 0.9212 |
0.0318 | 4.0 | 1632 | 0.1815 | 0.9696 | 0.9197 | 0.9087 | 0.9310 |
0.0098 | 5.0 | 2040 | 0.2013 | 0.9705 | 0.9227 | 0.9052 | 0.9409 |
0.0098 | 6.0 | 2448 | 0.2173 | 0.9733 | 0.9294 | 0.9183 | 0.9409 |
0.0031 | 7.0 | 2856 | 0.2324 | 0.9696 | 0.9189 | 0.9167 | 0.9212 |
0.0024 | 8.0 | 3264 | 0.2396 | 0.9714 | 0.9249 | 0.9095 | 0.9409 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3