bert-emotion
This model is a fine-tuned version of distilbert-base-cased on the tweet_eval dataset. It achieves the following results on the evaluation set:
- Loss: 1.2559
- Precision: 0.7221
- Recall: 0.7242
- Fscore: 0.7223
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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | Fscore |
|---|---|---|---|---|---|---|
| 0.8588 | 1.0 | 815 | 0.8342 | 0.7807 | 0.6117 | 0.6364 |
| 0.5394 | 2.0 | 1630 | 0.9126 | 0.7363 | 0.6923 | 0.7096 |
| 0.2805 | 3.0 | 2445 | 1.2559 | 0.7221 | 0.7242 | 0.7223 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2