MARBERT-QADI
This model was trained from scratch on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.0342
- Macro F1: 0.5099
- Accuracy: 0.5138
- Recall: 0.5136
- Precision: 0.6223
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: 4e-06
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Macro F1 | Accuracy | Recall | Precision |
|---|
| 0.8588 | 1.0 | 1125 | 0.7883 | 0.7550 | 0.7554 | 0.7552 | 0.7609 |
| 0.7475 | 2.0 | 2250 | 0.7718 | 0.7632 | 0.7634 | 0.7631 | 0.7653 |
| 0.6527 | 3.0 | 3375 | 0.7758 | 0.7668 | 0.7673 | 0.7671 | 0.7679 |
| 0.5654 | 4.0 | 4500 | 0.7845 | 0.7665 | 0.7673 | 0.7671 | 0.7682 |
| 0.5001 | 5.0 | 5625 | 0.8068 | 0.7650 | 0.7663 | 0.7660 | 0.7657 |
| 0.4641 | 6.0 | 6750 | 0.8216 | 0.7647 | 0.7658 | 0.7655 | 0.7650 |
| 0.4049 | 7.0 | 7875 | 0.8393 | 0.7645 | 0.7654 | 0.7649 | 0.7657 |
| 0.3773 | 8.0 | 9000 | 0.8477 | 0.7651 | 0.7657 | 0.7654 | 0.7659 |
| 0.3393 | 9.0 | 10125 | 0.8569 | 0.7663 | 0.7669 | 0.7665 | 0.7670 |
| 0.3383 | 10.0 | 11250 | 0.8589 | 0.7663 | 0.7669 | 0.7666 | 0.7667 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.19.1
