<!-- 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. -->
bertweet-base-Twitter_Sentiment_Analysis_v3
This model is a fine-tuned version of vinai/bertweet-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5141
- Accuracy: 0.8552
- Weighted f1: 0.8541
- Micro f1: 0.8552
- Macro f1: 0.8178
- Weighted recall: 0.8552
- Micro recall: 0.8552
- Macro recall: 0.8207
- Weighted precision: 0.8541
- Micro precision: 0.8552
- Macro precision: 0.8171
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 12
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Weighted f1 | Micro f1 | Macro f1 | Weighted recall | Micro recall | Macro recall | Weighted precision | Micro precision | Macro precision |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0.4404 | 1.0 | 149 | 0.2815 | 0.8314 | 0.8153 | 0.8314 | 0.7660 | 0.8314 | 0.8314 | 0.7560 | 0.8331 | 0.8314 | 0.8244 |
0.2493 | 2.0 | 298 | 0.2558 | 0.8588 | 0.8523 | 0.8588 | 0.8151 | 0.8588 | 0.8588 | 0.7961 | 0.8554 | 0.8588 | 0.8463 |
0.1905 | 3.0 | 447 | 0.2734 | 0.8580 | 0.8534 | 0.8580 | 0.8164 | 0.8580 | 0.8580 | 0.8094 | 0.8553 | 0.8580 | 0.8326 |
0.1504 | 4.0 | 596 | 0.3150 | 0.8631 | 0.8607 | 0.8631 | 0.8247 | 0.8631 | 0.8631 | 0.8291 | 0.8627 | 0.8631 | 0.8279 |
0.112 | 5.0 | 745 | 0.3451 | 0.8564 | 0.8522 | 0.8564 | 0.8145 | 0.8564 | 0.8564 | 0.8136 | 0.8544 | 0.8564 | 0.8254 |
0.0885 | 6.0 | 894 | 0.3929 | 0.8576 | 0.8532 | 0.8576 | 0.8158 | 0.8576 | 0.8576 | 0.8123 | 0.8554 | 0.8576 | 0.8293 |
0.0735 | 7.0 | 1043 | 0.4233 | 0.8564 | 0.8541 | 0.8564 | 0.8164 | 0.8564 | 0.8564 | 0.8128 | 0.8535 | 0.8564 | 0.8225 |
0.0642 | 8.0 | 1192 | 0.4454 | 0.8525 | 0.8495 | 0.8525 | 0.8106 | 0.8525 | 0.8525 | 0.8055 | 0.8491 | 0.8525 | 0.8192 |
0.0512 | 9.0 | 1341 | 0.5098 | 0.8537 | 0.8543 | 0.8537 | 0.8194 | 0.8537 | 0.8537 | 0.8261 | 0.8552 | 0.8537 | 0.8133 |
0.0448 | 10.0 | 1490 | 0.5268 | 0.8537 | 0.8538 | 0.8537 | 0.8170 | 0.8537 | 0.8537 | 0.8256 | 0.8549 | 0.8537 | 0.8101 |
0.038 | 11.0 | 1639 | 0.5076 | 0.8564 | 0.8555 | 0.8564 | 0.8195 | 0.8564 | 0.8564 | 0.8209 | 0.8551 | 0.8564 | 0.8191 |
0.0357 | 12.0 | 1788 | 0.5141 | 0.8552 | 0.8541 | 0.8552 | 0.8178 | 0.8552 | 0.8552 | 0.8207 | 0.8541 | 0.8552 | 0.8171 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.12.1
- Datasets 2.9.0
- Tokenizers 0.12.1