arabert_baseline_relevance_task7_fold0

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1556
• Qwk: 0.3529
• Mse: 0.1556

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse
No log	0.3333	2	0.3146	0.1946	0.3146
No log	0.6667	4	0.2014	0.5139	0.2014
No log	1.0	6	0.3227	0.2282	0.3227
No log	1.3333	8	0.3751	0.2780	0.3751
No log	1.6667	10	0.1512	0.4083	0.1512
No log	2.0	12	0.1822	0.5139	0.1822
No log	2.3333	14	0.2006	0.3836	0.2006
No log	2.6667	16	0.1882	0.1946	0.1882
No log	3.0	18	0.1835	0.1946	0.1835
No log	3.3333	20	0.1790	0.2925	0.1790
No log	3.6667	22	0.1666	0.3365	0.1666
No log	4.0	24	0.1570	0.4012	0.1570
No log	4.3333	26	0.1650	0.5139	0.1650
No log	4.6667	28	0.1468	0.3919	0.1468
No log	5.0	30	0.1342	0.3919	0.1342
No log	5.3333	32	0.1362	0.3289	0.1362
No log	5.6667	34	0.1387	0.3289	0.1387
No log	6.0	36	0.1474	0.3919	0.1474
No log	6.3333	38	0.1652	0.3919	0.1652
No log	6.6667	40	0.1889	0.4483	0.1889
No log	7.0	42	0.1978	0.4483	0.1978
No log	7.3333	44	0.1910	0.4483	0.1910
No log	7.6667	46	0.1747	0.2568	0.1747
No log	8.0	48	0.1647	0.4083	0.1647
No log	8.3333	50	0.1586	0.4083	0.1586
No log	8.6667	52	0.1552	0.3529	0.1552
No log	9.0	54	0.1544	0.3529	0.1544
No log	9.3333	56	0.1548	0.3529	0.1548
No log	9.6667	58	0.1554	0.3529	0.1554
No log	10.0	60	0.1556	0.3529	0.1556

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1