cs4248_albert-base-v2_bilstm_qa
This model is a fine-tuned version of albert-base-v2 on the squad dataset. It achieves the following results on the evaluation set:
- Loss: 4.9086
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: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 4248
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.4082 | 1.0 | 500 | 4.4031 |
| 3.951 | 2.0 | 1000 | 4.3329 |
| 3.6099 | 3.0 | 1500 | 4.3608 |
| 3.3651 | 4.0 | 2000 | 4.4929 |
| 3.0356 | 5.0 | 2500 | 4.5390 |
| 2.8111 | 6.0 | 3000 | 4.7661 |
| 2.5587 | 7.0 | 3500 | 4.8550 |
| 2.4021 | 8.0 | 4000 | 4.9086 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1