tiny-mlm-glue-wnli-target-glue-mrpc
This model is a fine-tuned version of muhtasham/tiny-mlm-glue-wnli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.1768
- Accuracy: 0.7230
- F1: 0.8094
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- training_steps: 5000
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.5908 | 4.35 | 500 | 0.5715 | 0.7083 | 0.8059 |
| 0.4649 | 8.7 | 1000 | 0.5978 | 0.7206 | 0.8106 |
| 0.3312 | 13.04 | 1500 | 0.6800 | 0.7255 | 0.8108 |
| 0.2207 | 17.39 | 2000 | 0.8000 | 0.7157 | 0.8014 |
| 0.1398 | 21.74 | 2500 | 0.9734 | 0.7255 | 0.8069 |
| 0.0984 | 26.09 | 3000 | 1.1768 | 0.7230 | 0.8094 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2