bert-base-arabertv2_D3Tok_EMD_19levels

This model is a fine-tuned version of aubmindlab/bert-base-arabertv2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0306
• Macro F1: 0.4521
• Macro Precision: 0.4818
• Macro Recall: 0.4453
• Accuracy: 0.5291
• Accuracy With Margin: 0.6966
• Distance: 1.1860
• Quadratic weighted kappa: 0.7940
• Accuracy 7: 0.6300
• Accuracy 5: 0.6726
• Accuracy 3: 0.7412

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: 5e-05
• train_batch_size: 64
• eval_batch_size: 16
• 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: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Macro F1	Macro Precision	Macro Recall	Accuracy	Accuracy With Margin	Distance	Quadratic weighted kappa	Accuracy 7	Accuracy 5	Accuracy 3
1.0454	1.0	857	0.9505	0.2627	0.3239	0.3181	0.4703	0.6428	1.3360	0.7758	0.6030	0.6647	0.7330
0.718	2.0	1714	0.8807	0.3647	0.3734	0.3709	0.5175	0.6841	1.1967	0.7981	0.6297	0.6787	0.7490
0.5604	3.0	2571	0.9115	0.4027	0.4449	0.3915	0.5353	0.6837	1.1989	0.7920	0.6305	0.6746	0.7393
0.4615	4.0	3428	0.9439	0.4163	0.4850	0.4124	0.5328	0.6830	1.2074	0.7942	0.6256	0.6699	0.7378
0.3547	5.0	4285	0.9803	0.4059	0.4732	0.3965	0.5215	0.6807	1.2193	0.7864	0.6218	0.6655	0.7363
0.2972	6.0	5142	0.9809	0.4499	0.4851	0.4425	0.5356	0.6856	1.2003	0.7931	0.6287	0.6695	0.7393
0.2529	7.0	5999	0.9922	0.4427	0.4791	0.4293	0.5280	0.6885	1.1907	0.7946	0.6268	0.6715	0.7431
0.2002	8.0	6856	1.0047	0.4516	0.4747	0.4448	0.5316	0.6902	1.1818	0.7974	0.6306	0.6731	0.7390
0.1734	9.0	7713	1.0181	0.4545	0.4949	0.4515	0.5353	0.6951	1.1945	0.7920	0.6354	0.6784	0.7440
0.1466	10.0	8570	1.0184	0.4469	0.4748	0.4414	0.5271	0.6951	1.1871	0.7960	0.6312	0.6739	0.7404
0.1342	11.0	9427	1.0291	0.4496	0.4739	0.4471	0.5287	0.6985	1.1815	0.7980	0.6328	0.6770	0.7446
0.1166	12.0	10284	1.0306	0.4521	0.4818	0.4453	0.5291	0.6966	1.1860	0.7940	0.6300	0.6726	0.7412

Framework versions

• Transformers 4.53.2
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.21.2