small-mlm-glue-mrpc-custom-tokenizer-target-glue-qqp
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: 0.4321
- Accuracy: 0.7877
- F1: 0.7536
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 |
|---|---|---|---|---|---|
| 0.5322 | 0.04 | 500 | 0.5216 | 0.7117 | 0.6856 |
| 0.4868 | 0.09 | 1000 | 0.4842 | 0.7449 | 0.7042 |
| 0.4808 | 0.13 | 1500 | 0.4606 | 0.7593 | 0.7161 |
| 0.4628 | 0.18 | 2000 | 0.4532 | 0.7623 | 0.7245 |
| 0.4567 | 0.22 | 2500 | 0.4799 | 0.7495 | 0.7268 |
| 0.4367 | 0.26 | 3000 | 0.4231 | 0.7917 | 0.7319 |
| 0.4362 | 0.31 | 3500 | 0.4171 | 0.8017 | 0.7287 |
| 0.4261 | 0.35 | 4000 | 0.4317 | 0.7814 | 0.7476 |
| 0.4268 | 0.4 | 4500 | 0.4240 | 0.7867 | 0.7519 |
| 0.413 | 0.44 | 5000 | 0.4321 | 0.7877 | 0.7536 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2