small-mlm-glue-mrpc-custom-tokenizer-target-glue-qnli
This model is a fine-tuned version of muhtasham/small-mlm-glue-mrpc-custom-tokenizer on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4493
- Accuracy: 0.7986
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 |
|---|---|---|---|---|
| 0.5672 | 0.15 | 500 | 0.4950 | 0.7650 |
| 0.532 | 0.31 | 1000 | 0.4894 | 0.7710 |
| 0.5191 | 0.46 | 1500 | 0.5007 | 0.7681 |
| 0.5102 | 0.61 | 2000 | 0.4682 | 0.7873 |
| 0.5033 | 0.76 | 2500 | 0.4596 | 0.7897 |
| 0.4975 | 0.92 | 3000 | 0.4537 | 0.7908 |
| 0.4751 | 1.07 | 3500 | 0.4637 | 0.7900 |
| 0.4547 | 1.22 | 4000 | 0.5252 | 0.7717 |
| 0.45 | 1.37 | 4500 | 0.4494 | 0.8003 |
| 0.454 | 1.53 | 5000 | 0.4493 | 0.7986 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2