bert_emo_classifier
This model is a fine-tuned version of bert-base-uncased on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.2724
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: 4
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.9319 | 0.25 | 500 | 0.4107 |
0.3265 | 0.5 | 1000 | 0.3068 |
0.2458 | 0.75 | 1500 | 0.2721 |
0.2487 | 1.0 | 2000 | 0.2313 |
0.158 | 1.25 | 2500 | 0.2422 |
0.1796 | 1.5 | 3000 | 0.2162 |
0.145 | 1.75 | 3500 | 0.1951 |
0.1648 | 2.0 | 4000 | 0.1908 |
0.1048 | 2.25 | 4500 | 0.2399 |
0.1171 | 2.5 | 5000 | 0.2230 |
0.1116 | 2.75 | 5500 | 0.2244 |
0.1122 | 3.0 | 6000 | 0.2250 |
0.0713 | 3.25 | 6500 | 0.2616 |
0.0697 | 3.5 | 7000 | 0.2672 |
0.0775 | 3.75 | 7500 | 0.2748 |
0.0742 | 4.0 | 8000 | 0.2724 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
- Downloads last month
- 9