<!-- 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. -->
hBERTv1_new_pretrain_48_KD_cola
This model is a fine-tuned version of gokuls/bert_12_layer_model_v1_complete_training_new_48_KD on the GLUE COLA dataset. It achieves the following results on the evaluation set:
- Loss: 0.6163
- Matthews Correlation: 0.0
- Accuracy: 0.6913
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 | Matthews Correlation | Accuracy |
|---|---|---|---|---|---|
| 0.6281 | 1.0 | 67 | 0.6241 | 0.0 | 0.6913 |
| 0.6172 | 2.0 | 134 | 0.6471 | 0.0 | 0.6913 |
| 0.6123 | 3.0 | 201 | 0.6185 | 0.0 | 0.6913 |
| 0.6138 | 4.0 | 268 | 0.6180 | 0.0 | 0.6913 |
| 0.6029 | 5.0 | 335 | 0.6170 | 0.0021 | 0.6846 |
| 0.5795 | 6.0 | 402 | 0.6163 | 0.0 | 0.6913 |
| 0.556 | 7.0 | 469 | 0.6529 | 0.0735 | 0.6328 |
| 0.5273 | 8.0 | 536 | 0.6502 | 0.1135 | 0.6510 |
| 0.5108 | 9.0 | 603 | 0.7095 | 0.1134 | 0.6318 |
| 0.4911 | 10.0 | 670 | 0.6667 | 0.0454 | 0.6644 |
| 0.4768 | 11.0 | 737 | 0.6869 | 0.0658 | 0.6232 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3
