<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
bert-base-cased-stsb
This model is a fine-tuned version of bert-base-cased on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 0.4322
- Pearson: 0.9007
- Spearmanr: 0.8963
- Combined Score: 0.8985
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 10.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Pearson | Spearmanr | Combined Score |
|---|---|---|---|---|---|---|
| 1.6464 | 1.39 | 500 | 0.5662 | 0.8820 | 0.8814 | 0.8817 |
| 0.3329 | 2.78 | 1000 | 0.5070 | 0.8913 | 0.8883 | 0.8898 |
| 0.173 | 4.17 | 1500 | 0.4465 | 0.8988 | 0.8943 | 0.8966 |
| 0.1085 | 5.56 | 2000 | 0.4537 | 0.8958 | 0.8917 | 0.8937 |
| 0.0816 | 6.94 | 2500 | 0.4594 | 0.8977 | 0.8933 | 0.8955 |
| 0.0621 | 8.33 | 3000 | 0.4450 | 0.8997 | 0.8950 | 0.8974 |
| 0.0519 | 9.72 | 3500 | 0.4322 | 0.9007 | 0.8963 | 0.8985 |
Framework versions
- Transformers 4.21.3
- Pytorch 1.7.1
- Datasets 1.18.3
- Tokenizers 0.11.6
