mobilebert_sa_GLUE_Experiment_logit_kd_mrpc_128
This model is a fine-tuned version of google/mobilebert-uncased on the GLUE MRPC dataset. It achieves the following results on the evaluation set:
- Loss: 0.5213
- Accuracy: 0.6740
- F1: 0.7787
- Combined Score: 0.7264
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: 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 | Accuracy | F1 | Combined Score |
|---|---|---|---|---|---|---|
| 0.6368 | 1.0 | 29 | 0.5564 | 0.6838 | 0.8122 | 0.7480 |
| 0.6099 | 2.0 | 58 | 0.5557 | 0.6838 | 0.8122 | 0.7480 |
| 0.611 | 3.0 | 87 | 0.5555 | 0.6838 | 0.8122 | 0.7480 |
| 0.6101 | 4.0 | 116 | 0.5568 | 0.6838 | 0.8122 | 0.7480 |
| 0.608 | 5.0 | 145 | 0.5540 | 0.6838 | 0.8122 | 0.7480 |
| 0.6037 | 6.0 | 174 | 0.5492 | 0.6838 | 0.8122 | 0.7480 |
| 0.5761 | 7.0 | 203 | 0.6065 | 0.6103 | 0.6851 | 0.6477 |
| 0.4782 | 8.0 | 232 | 0.5341 | 0.6863 | 0.7801 | 0.7332 |
| 0.4111 | 9.0 | 261 | 0.5213 | 0.6740 | 0.7787 | 0.7264 |
| 0.3526 | 10.0 | 290 | 0.5792 | 0.6863 | 0.7867 | 0.7365 |
| 0.3188 | 11.0 | 319 | 0.5760 | 0.6936 | 0.7764 | 0.7350 |
| 0.2918 | 12.0 | 348 | 0.6406 | 0.6912 | 0.7879 | 0.7395 |
| 0.2568 | 13.0 | 377 | 0.5908 | 0.6765 | 0.7537 | 0.7151 |
| 0.2472 | 14.0 | 406 | 0.5966 | 0.6863 | 0.7664 | 0.7263 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2