tiny-mlm-snli-target-glue-qqp
This model is a fine-tuned version of muhtasham/tiny-mlm-snli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4222
- Accuracy: 0.7880
- F1: 0.7489
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 |
|---|---|---|---|---|---|
| 0.5788 | 0.04 | 500 | 0.5176 | 0.7289 | 0.6776 |
| 0.5105 | 0.09 | 1000 | 0.4825 | 0.7519 | 0.7012 |
| 0.4971 | 0.13 | 1500 | 0.4913 | 0.7412 | 0.7140 |
| 0.4815 | 0.18 | 2000 | 0.4669 | 0.7540 | 0.7200 |
| 0.4726 | 0.22 | 2500 | 0.4566 | 0.7630 | 0.7273 |
| 0.4572 | 0.26 | 3000 | 0.4494 | 0.7682 | 0.7348 |
| 0.4579 | 0.31 | 3500 | 0.4467 | 0.7694 | 0.7384 |
| 0.451 | 0.35 | 4000 | 0.4459 | 0.7696 | 0.7400 |
| 0.4519 | 0.4 | 4500 | 0.4442 | 0.7686 | 0.7411 |
| 0.4417 | 0.44 | 5000 | 0.4222 | 0.7880 | 0.7489 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2