bert-reg-crossencoder-mse
This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0752
- Mse: 0.0752
- Mae: 0.2120
- Pearson Corr: 0.3937
- Spearman Corr: 0.3178
- Cosine Sim: 0.9163
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 7
Training results
| Training Loss | Epoch | Step | Validation Loss | Mse | Mae | Pearson Corr | Spearman Corr | Cosine Sim |
|---|---|---|---|---|---|---|---|---|
| 0.1034 | 1.0 | 41 | 0.0704 | 0.0704 | 0.2198 | 0.1914 | 0.2429 | 0.9070 |
| 0.097 | 2.0 | 82 | 0.0739 | 0.0739 | 0.2161 | 0.2185 | 0.2208 | 0.9059 |
| 0.0877 | 3.0 | 123 | 0.0663 | 0.0663 | 0.2154 | 0.3214 | 0.2426 | 0.9133 |
| 0.0679 | 4.0 | 164 | 0.0723 | 0.0723 | 0.2054 | 0.3722 | 0.3382 | 0.9175 |
| 0.0569 | 5.0 | 205 | 0.0644 | 0.0644 | 0.2058 | 0.3867 | 0.3552 | 0.9155 |
| 0.0408 | 6.0 | 246 | 0.0773 | 0.0773 | 0.2102 | 0.4045 | 0.3105 | 0.9190 |
| 0.0317 | 7.0 | 287 | 0.0752 | 0.0752 | 0.2120 | 0.3937 | 0.3178 | 0.9163 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.5.0+cu121
- Datasets 3.1.0
- Tokenizers 0.19.1