viettel-videberta-finetune-viquad-model6
This model is a fine-tuned version of Fsoft-AIC/videberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.5762
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: 14
- eval_batch_size: 14
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.7932 | 0.58 | 800 | 3.3482 |
| 2.8701 | 1.15 | 1600 | 2.8881 |
| 2.5971 | 1.73 | 2400 | 2.5908 |
| 2.2575 | 2.31 | 3200 | 2.5129 |
| 2.0777 | 2.88 | 4000 | 2.3878 |
| 1.8617 | 3.46 | 4800 | 2.4493 |
| 1.7606 | 4.04 | 5600 | 2.4275 |
| 1.5691 | 4.61 | 6400 | 2.5150 |
| 1.4966 | 5.19 | 7200 | 2.5777 |
| 1.351 | 5.77 | 8000 | 2.5762 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3