bert-base-cased-english-sentweet-Derogatory
This model is a fine-tuned version of bert-base-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9504
- Accuracy: 0.7917
- Precision: 0.7961
- Recall: 0.7982
- 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 | 0.4578 | 0.8160 | 0.8363 | 0.8289 | 0.8157 |
| No log | 2.0 | 162 | 0.4759 | 0.8264 | 0.8459 | 0.8391 | 0.8262 |
| No log | 3.0 | 243 | 0.5410 | 0.8160 | 0.8317 | 0.8275 | 0.8159 |
| No log | 4.0 | 324 | 0.5789 | 0.7569 | 0.7545 | 0.7557 | 0.7549 |
| No log | 5.0 | 405 | 0.9143 | 0.7847 | 0.7891 | 0.7911 | 0.7846 |
| No log | 6.0 | 486 | 0.9504 | 0.7917 | 0.7961 | 0.7982 | 0.7916 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.1+cu117
- Datasets 2.6.1
- Tokenizers 0.11.0