arabert_baseline_relevance_task1_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: 0.1576
• Qwk: 0.2222
• Mse: 0.1601

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	0.9197	0.0278	0.9116
No log	0.6667	4	0.1428	0.2105	0.1467
No log	1.0	6	0.1823	0.3467	0.1888
No log	1.3333	8	0.1817	0.0808	0.1888
No log	1.6667	10	0.2156	0.0	0.2206
No log	2.0	12	0.1825	0.0	0.1849
No log	2.3333	14	0.1725	0.0392	0.1727
No log	2.6667	16	0.2068	0.0392	0.2060
No log	3.0	18	0.1522	0.0808	0.1531
No log	3.3333	20	0.1588	0.1250	0.1623
No log	3.6667	22	0.1635	0.1250	0.1682
No log	4.0	24	0.1609	0.0808	0.1658
No log	4.3333	26	0.1526	0.0808	0.1565
No log	4.6667	28	0.1503	0.0392	0.1529
No log	5.0	30	0.1548	0.0392	0.1558
No log	5.3333	32	0.1602	0.0808	0.1608
No log	5.6667	34	0.1480	0.1720	0.1493
No log	6.0	36	0.1509	0.2759	0.1526
No log	6.3333	38	0.1504	0.3333	0.1525
No log	6.6667	40	0.1502	0.3333	0.1526
No log	7.0	42	0.1518	0.1720	0.1539
No log	7.3333	44	0.1602	0.1250	0.1617
No log	7.6667	46	0.1798	0.1250	0.1806
No log	8.0	48	0.1921	0.2364	0.1927
No log	8.3333	50	0.1893	0.2364	0.1900
No log	8.6667	52	0.1769	0.2222	0.1778
No log	9.0	54	0.1663	0.2759	0.1678
No log	9.3333	56	0.1609	0.2759	0.1629
No log	9.6667	58	0.1584	0.2222	0.1608
No log	10.0	60	0.1576	0.2222	0.1601

Framework versions

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