add_BERT_no_pretrain_stsb
This model is a fine-tuned version of on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 2.3307
- Pearson: 0.0719
- Spearmanr: 0.0575
- Combined Score: 0.0647
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: 4e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Pearson | Spearmanr | Combined Score |
|---|---|---|---|---|---|---|
| 2.4937 | 1.0 | 45 | 2.6373 | 0.0410 | 0.0287 | 0.0348 |
| 2.22 | 2.0 | 90 | 2.4288 | 0.0740 | 0.0586 | 0.0663 |
| 2.1554 | 3.0 | 135 | 2.3869 | 0.0609 | 0.0498 | 0.0554 |
| 2.0556 | 4.0 | 180 | 2.5740 | 0.0800 | 0.0717 | 0.0759 |
| 2.0221 | 5.0 | 225 | 2.4656 | 0.0849 | 0.0654 | 0.0752 |
| 2.0403 | 6.0 | 270 | 2.3307 | 0.0719 | 0.0575 | 0.0647 |
| 2.1732 | 7.0 | 315 | 2.5174 | 0.0699 | 0.0584 | 0.0641 |
| 2.0399 | 8.0 | 360 | 2.5648 | 0.0718 | 0.0605 | 0.0662 |
| 2.0765 | 9.0 | 405 | 2.3373 | 0.0621 | 0.0491 | 0.0556 |
| 2.0538 | 10.0 | 450 | 2.6402 | 0.0463 | 0.0431 | 0.0447 |
| 2.0147 | 11.0 | 495 | 2.4727 | 0.0540 | 0.0471 | 0.0506 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3