<!-- 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. -->
full-review-clf
This model is a fine-tuned version of cardiffnlp/twitter-roberta-base-sentiment-latest on justina/yelp-boba-reviews dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8198
- F1 Macro: 0.6358
- Aucpr Macro: 0.6658
- Accuracy: 0.7185
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 Macro | Aucpr Macro | Accuracy |
|---|---|---|---|---|---|---|
| 0.723 | 0.43 | 500 | 0.7576 | 0.5979 | 0.6652 | 0.6831 |
| 0.7307 | 0.87 | 1000 | 0.6862 | 0.6368 | 0.6752 | 0.7185 |
| 0.5828 | 1.3 | 1500 | 0.7398 | 0.6439 | 0.6661 | 0.7255 |
| 0.6236 | 1.73 | 2000 | 0.7878 | 0.6212 | 0.6690 | 0.7069 |
| 0.3739 | 2.16 | 2500 | 0.8138 | 0.6447 | 0.6752 | 0.7170 |
| 0.4235 | 2.6 | 3000 | 0.8048 | 0.6490 | 0.6673 | 0.7255 |
| 0.3684 | 3.03 | 3500 | 0.9615 | 0.6483 | 0.6715 | 0.7205 |
| 0.3243 | 3.46 | 4000 | 1.0931 | 0.6432 | 0.6632 | 0.7235 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3
