debertav3-qqp
This model is a fine-tuned version of microsoft/deberta-v3-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2065
- Accuracy: 0.9217
- F1: 0.8962
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: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 0.1
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| No log | 0 | 0 | 0.7186 | 0.3682 | 0.5382 |
| 0.3736 | 0.1759 | 1000 | 0.3048 | 0.8596 | 0.8311 |
| 0.2823 | 0.3517 | 2000 | 0.2573 | 0.8945 | 0.8573 |
| 0.2565 | 0.5276 | 3000 | 0.2431 | 0.8960 | 0.8627 |
| 0.2399 | 0.7035 | 4000 | 0.2236 | 0.9048 | 0.8728 |
| 0.2322 | 0.8794 | 5000 | 0.2188 | 0.9108 | 0.8835 |
| 0.2059 | 1.0552 | 6000 | 0.2285 | 0.9137 | 0.8838 |
| 0.1626 | 1.2311 | 7000 | 0.2297 | 0.9169 | 0.8893 |
| 0.1568 | 1.4070 | 8000 | 0.2105 | 0.9160 | 0.8886 |
| 0.1533 | 1.5828 | 9000 | 0.2124 | 0.9194 | 0.8932 |
| 0.1501 | 1.7587 | 10000 | 0.2118 | 0.9206 | 0.8948 |
| 0.1459 | 1.9346 | 11000 | 0.2074 | 0.9211 | 0.8946 |
| 0.1459 | 2.0 | 11372 | 0.2065 | 0.9217 | 0.8962 |
Framework versions
- Transformers 5.20.2
- Pytorch 2.11.0+cu130
- Datasets 5.1.0
- Tokenizers 0.23.3