albert-base-v2-finetuned-ours-DS
This model is a fine-tuned version of albert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.8481
- Accuracy: 0.665
- Precision: 0.6202
- Recall: 0.6137
- F1: 0.5973
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: 1.0031150633640573e-05
- train_batch_size: 8
- eval_batch_size: 16
- seed: 43
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 1.053 | 0.49 | 99 | 1.0208 | 0.56 | 0.5650 | 0.4382 | 0.3908 |
| 0.8912 | 0.99 | 198 | 0.8648 | 0.59 | 0.3832 | 0.5279 | 0.4374 |
| 0.777 | 1.48 | 297 | 0.7986 | 0.665 | 0.6579 | 0.5985 | 0.5704 |
| 0.7391 | 1.98 | 396 | 0.8363 | 0.58 | 0.6389 | 0.5669 | 0.4966 |
| 0.5861 | 2.48 | 495 | 0.8394 | 0.685 | 0.6380 | 0.6313 | 0.6312 |
| 0.5711 | 2.97 | 594 | 0.9090 | 0.65 | 0.6102 | 0.5833 | 0.5900 |
| 0.4296 | 3.46 | 693 | 0.8481 | 0.665 | 0.6202 | 0.6137 | 0.5973 |
Framework versions
- Transformers 4.21.3
- Pytorch 1.12.1+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1