<!-- 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_add_GLUE_Experiment_logit_kd_mrpc
This model is a fine-tuned version of distilbert-base-uncased on the GLUE MRPC dataset. It achieves the following results on the evaluation set:
- Loss: 0.5207
- Accuracy: 0.3162
- F1: 0.0
- Combined Score: 0.1581
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 |
---|---|---|---|---|---|---|
0.564 | 1.0 | 15 | 0.5300 | 0.3162 | 0.0 | 0.1581 |
0.533 | 2.0 | 30 | 0.5323 | 0.3162 | 0.0 | 0.1581 |
0.5302 | 3.0 | 45 | 0.5290 | 0.3162 | 0.0 | 0.1581 |
0.5312 | 4.0 | 60 | 0.5289 | 0.3162 | 0.0 | 0.1581 |
0.527 | 5.0 | 75 | 0.5306 | 0.3162 | 0.0 | 0.1581 |
0.5229 | 6.0 | 90 | 0.5207 | 0.3162 | 0.0 | 0.1581 |
0.5088 | 7.0 | 105 | 0.5358 | 0.5539 | 0.5806 | 0.5673 |
0.5003 | 8.0 | 120 | 0.5299 | 0.4902 | 0.4611 | 0.4757 |
0.4825 | 9.0 | 135 | 0.5323 | 0.3627 | 0.1824 | 0.2726 |
0.4628 | 10.0 | 150 | 0.5373 | 0.5196 | 0.5377 | 0.5287 |
0.451 | 11.0 | 165 | 0.5513 | 0.5417 | 0.5854 | 0.5635 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2