qa-persian-roberta-fa-zwnj-base
This model is a fine-tuned version of makhataei/qa-persian-roberta-fa-zwnj-base on the parsinlu_reading_comprehension dataset. It achieves the following results on the evaluation set:
- Loss: 5.6923
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: 0.001
- train_batch_size: 14
- eval_batch_size: 14
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 5.2326 | 1.0 | 44 | 5.6923 |
| 5.3984 | 2.0 | 88 | 5.6923 |
| 5.4673 | 3.0 | 132 | 5.6923 |
| 5.5463 | 4.0 | 176 | 5.6923 |
| 5.7433 | 5.0 | 220 | 5.6923 |
| 5.7994 | 6.0 | 264 | 5.6923 |
| 5.8169 | 7.0 | 308 | 5.6923 |
| 5.8149 | 8.0 | 352 | 5.6923 |
| 5.8499 | 9.0 | 396 | 5.6923 |
| 5.889 | 10.0 | 440 | 5.6923 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.0.1+cu117
- Datasets 2.15.0
- Tokenizers 0.15.0