bert-base-uncased-english-sentweet-derogatory
This model is a fine-tuned version of nlptown/bert-base-multilingual-uncased-sentiment on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.1640
- Accuracy: 0.7917
- Precision: 0.8058
- Recall: 0.8025
- F1: 0.7916
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 81 | 0.4757 | 0.8021 | 0.8300 | 0.8171 | 0.8014 |
| No log | 2.0 | 162 | 0.5035 | 0.8194 | 0.8412 | 0.8328 | 0.8191 |
| No log | 3.0 | 243 | 0.5446 | 0.8021 | 0.8220 | 0.8149 | 0.8018 |
| No log | 4.0 | 324 | 0.7602 | 0.7465 | 0.7482 | 0.7507 | 0.7462 |
| No log | 5.0 | 405 | 1.0083 | 0.7743 | 0.7793 | 0.7810 | 0.7742 |
| No log | 6.0 | 486 | 1.1640 | 0.7917 | 0.8058 | 0.8025 | 0.7916 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.1+cu117
- Datasets 2.6.1
- Tokenizers 0.11.0