<!-- 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. -->
distilbert_sa_GLUE_Experiment_logit_kd_data_aug_qqp
This model is a fine-tuned version of distilbert-base-uncased on the GLUE QQP dataset. It achieves the following results on the evaluation set:
- Loss: 0.6508
- Accuracy: 0.6562
- F1: 0.1361
- Combined Score: 0.3962
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: 5e-05
- train_batch_size: 256
- eval_batch_size: 256
- 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
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Combined Score |
---|---|---|---|---|---|---|
0.7744 | 1.0 | 29671 | 0.7100 | 0.6422 | 0.0598 | 0.3510 |
0.6803 | 2.0 | 59342 | 0.6914 | 0.6548 | 0.1291 | 0.3919 |
0.6617 | 3.0 | 89013 | 0.6944 | 0.6588 | 0.1498 | 0.4043 |
0.6531 | 4.0 | 118684 | 0.6699 | 0.6632 | 0.1741 | 0.4187 |
0.6482 | 5.0 | 148355 | 0.6633 | 0.6622 | 0.1666 | 0.4144 |
0.6451 | 6.0 | 178026 | 0.6508 | 0.6562 | 0.1361 | 0.3962 |
0.6431 | 7.0 | 207697 | 0.6632 | 0.6595 | 0.1526 | 0.4060 |
0.6416 | 8.0 | 237368 | 0.6621 | 0.6634 | 0.1720 | 0.4177 |
0.6404 | 9.0 | 267039 | 0.6579 | 0.6691 | 0.2001 | 0.4346 |
0.6395 | 10.0 | 296710 | 0.6554 | 0.6690 | 0.2029 | 0.4359 |
0.6387 | 11.0 | 326381 | 0.6558 | 0.6637 | 0.1755 | 0.4196 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2