bert-eou-classifier_teacher

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.1555
• Accuracy: 0.791
• Auc: 0.865

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: 0.0002
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Auc
0.5875	1.0	622	0.4866	0.75	0.845
0.4703	2.0	1244	0.5337	0.76	0.855
0.4097	3.0	1866	0.5273	0.785	0.869
0.3598	4.0	2488	0.5383	0.795	0.868
0.3278	5.0	3110	0.6127	0.803	0.878
0.3019	6.0	3732	0.6487	0.804	0.878
0.2616	7.0	4354	0.7659	0.801	0.874
0.2451	8.0	4976	0.8012	0.793	0.871
0.2241	9.0	5598	0.8936	0.802	0.87
0.2044	10.0	6220	0.9513	0.8	0.869
0.2015	11.0	6842	0.9689	0.802	0.869
0.1834	12.0	7464	0.9756	0.799	0.869
0.1731	13.0	8086	0.9917	0.796	0.866
0.1455	14.0	8708	1.0958	0.794	0.863
0.1557	15.0	9330	1.0042	0.796	0.869
0.1316	16.0	9952	1.0996	0.796	0.865
0.1335	17.0	10574	1.2024	0.794	0.863
0.1201	18.0	11196	1.1508	0.791	0.865
0.1204	19.0	11818	1.1580	0.798	0.865
0.1137	20.0	12440	1.1555	0.791	0.865

Framework versions

• PEFT 0.18.0
• Transformers 4.57.3
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.22.1