<!-- 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. -->
add_BERT_48_stsb
This model is a fine-tuned version of gokuls/add_bert_12_layer_model_complete_training_new_48 on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 1.7022
- Pearson: 0.5286
- Spearmanr: 0.5238
- Combined Score: 0.5262
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 |
---|---|---|---|---|---|---|
2.3047 | 1.0 | 45 | 2.9668 | 0.0943 | 0.0857 | 0.0900 |
2.1213 | 2.0 | 90 | 2.5339 | 0.1279 | 0.0928 | 0.1104 |
1.9666 | 3.0 | 135 | 2.2189 | 0.2727 | 0.2634 | 0.2681 |
1.636 | 4.0 | 180 | 2.6526 | 0.3479 | 0.3438 | 0.3459 |
1.1382 | 5.0 | 225 | 2.1790 | 0.4250 | 0.4209 | 0.4229 |
0.7856 | 6.0 | 270 | 2.3985 | 0.4820 | 0.5071 | 0.4946 |
0.5806 | 7.0 | 315 | 1.7992 | 0.5140 | 0.5093 | 0.5117 |
0.4372 | 8.0 | 360 | 1.7022 | 0.5286 | 0.5238 | 0.5262 |
0.3305 | 9.0 | 405 | 2.1792 | 0.5342 | 0.5367 | 0.5355 |
0.2799 | 10.0 | 450 | 1.7458 | 0.5254 | 0.5218 | 0.5236 |
0.2292 | 11.0 | 495 | 1.8574 | 0.5459 | 0.5466 | 0.5462 |
0.1945 | 12.0 | 540 | 1.7717 | 0.5571 | 0.5572 | 0.5571 |
0.1809 | 13.0 | 585 | 2.2290 | 0.5158 | 0.5209 | 0.5184 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.0
- Tokenizers 0.13.3