<!-- 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. -->
sa_BERT_no_pretrain_stsb
This model is a fine-tuned version of bert-base-uncased on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 2.5396
- Pearson: 0.1394
- Spearmanr: 0.1246
- Combined Score: 0.1320
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: 96
- eval_batch_size: 96
- 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 |
|---|---|---|---|---|---|---|
| 2.257 | 1.0 | 60 | 3.1111 | 0.0528 | 0.0709 | 0.0619 |
| 2.0476 | 2.0 | 120 | 2.5396 | 0.1394 | 0.1246 | 0.1320 |
| 1.8905 | 3.0 | 180 | 2.5928 | 0.1553 | 0.1593 | 0.1573 |
| 1.5383 | 4.0 | 240 | 3.1130 | 0.1930 | 0.2086 | 0.2008 |
| 1.3384 | 5.0 | 300 | 2.8651 | 0.1788 | 0.2014 | 0.1901 |
| 1.1299 | 6.0 | 360 | 2.9651 | 0.1818 | 0.1947 | 0.1883 |
| 1.0952 | 7.0 | 420 | 2.6404 | 0.2100 | 0.2124 | 0.2112 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3
