metadata
library_name: peft
license: apache-2.0
base_model: Qwen/Qwen2.5-0.5B
tags:
- base_model:adapter:Qwen/Qwen2.5-0.5B
- lora
- transformers
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: saudi-eou-qwen-classifier
results: []
saudi-eou-qwen-classifier
This model is a fine-tuned version of Qwen/Qwen2.5-0.5B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5642
- Accuracy: 0.737
- Auc: 0.819
- F1: 0.694
- Precision: 0.649
- Recall: 0.746
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Auc | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|---|
| 1.7739 | 1.0 | 311 | 0.8898 | 0.623 | 0.683 | 0.58 | 0.523 | 0.653 |
| 0.726 | 2.0 | 622 | 0.6793 | 0.701 | 0.77 | 0.673 | 0.597 | 0.771 |
| 0.5642 | 3.0 | 933 | 0.6046 | 0.72 | 0.791 | 0.68 | 0.626 | 0.746 |
| 0.5018 | 4.0 | 1244 | 0.5779 | 0.728 | 0.804 | 0.685 | 0.637 | 0.74 |
| 0.4623 | 5.0 | 1555 | 0.5831 | 0.721 | 0.811 | 0.691 | 0.62 | 0.78 |
| 0.4352 | 6.0 | 1866 | 0.5661 | 0.735 | 0.815 | 0.694 | 0.643 | 0.754 |
| 0.415 | 7.0 | 2177 | 0.5631 | 0.739 | 0.818 | 0.696 | 0.651 | 0.749 |
| 0.3993 | 8.0 | 2488 | 0.5638 | 0.742 | 0.818 | 0.7 | 0.653 | 0.754 |
| 0.3891 | 9.0 | 2799 | 0.5646 | 0.739 | 0.819 | 0.697 | 0.65 | 0.751 |
| 0.3809 | 10.0 | 3110 | 0.5642 | 0.737 | 0.819 | 0.694 | 0.649 | 0.746 |
Framework versions
- PEFT 0.18.0
- Transformers 4.57.3
- Pytorch 2.9.0+cu126
- Datasets 4.4.1
- Tokenizers 0.22.1