small-mlm-glue-qqp-target-glue-qqp
This model is a fine-tuned version of muhtasham/small-mlm-glue-qqp on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3299
- Accuracy: 0.8510
- F1: 0.8133
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 |
|---|---|---|---|---|---|
| 0.469 | 0.04 | 500 | 0.4282 | 0.7881 | 0.7509 |
| 0.4137 | 0.09 | 1000 | 0.3791 | 0.8200 | 0.7692 |
| 0.3948 | 0.13 | 1500 | 0.3707 | 0.8244 | 0.7757 |
| 0.3793 | 0.18 | 2000 | 0.3735 | 0.8231 | 0.7883 |
| 0.3748 | 0.22 | 2500 | 0.3550 | 0.8377 | 0.7968 |
| 0.3578 | 0.26 | 3000 | 0.3466 | 0.8385 | 0.8006 |
| 0.3563 | 0.31 | 3500 | 0.3277 | 0.8526 | 0.8081 |
| 0.351 | 0.35 | 4000 | 0.3300 | 0.8493 | 0.8085 |
| 0.354 | 0.4 | 4500 | 0.3230 | 0.8543 | 0.8115 |
| 0.3413 | 0.44 | 5000 | 0.3299 | 0.8510 | 0.8133 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2