arabert_cross_organization_task3_fold5

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: 0.4649
• Qwk: 0.6602
• Mse: 0.4652

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: 64
• eval_batch_size: 64
• 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	0.125	2	1.3614	0.1417	1.3607
No log	0.25	4	0.9919	0.1465	0.9922
No log	0.375	6	0.7852	0.5086	0.7857
No log	0.5	8	0.8360	0.6182	0.8371
No log	0.625	10	0.6409	0.7584	0.6421
No log	0.75	12	0.5760	0.6105	0.5765
No log	0.875	14	0.5769	0.7460	0.5780
No log	1.0	16	0.5277	0.7247	0.5287
No log	1.125	18	0.5309	0.7698	0.5319
No log	1.25	20	0.6352	0.7920	0.6365
No log	1.375	22	0.4636	0.6839	0.4644
No log	1.5	24	0.4773	0.7038	0.4782
No log	1.625	26	0.5174	0.7610	0.5185
No log	1.75	28	0.5482	0.7835	0.5494
No log	1.875	30	0.4822	0.7417	0.4830
No log	2.0	32	0.4561	0.6787	0.4566
No log	2.125	34	0.4716	0.7343	0.4723
No log	2.25	36	0.5650	0.7930	0.5660
No log	2.375	38	0.4723	0.7646	0.4729
No log	2.5	40	0.4662	0.6525	0.4664
No log	2.625	42	0.4420	0.7033	0.4425
No log	2.75	44	0.5038	0.7717	0.5047
No log	2.875	46	0.5557	0.7833	0.5567
No log	3.0	48	0.4661	0.7378	0.4667
No log	3.125	50	0.4772	0.6547	0.4774
No log	3.25	52	0.4713	0.6765	0.4715
No log	3.375	54	0.4755	0.7516	0.4761
No log	3.5	56	0.4959	0.7626	0.4966
No log	3.625	58	0.4637	0.7460	0.4642
No log	3.75	60	0.4762	0.7443	0.4768
No log	3.875	62	0.4669	0.7437	0.4674
No log	4.0	64	0.4555	0.7387	0.4559
No log	4.125	66	0.4440	0.7154	0.4444
No log	4.25	68	0.4456	0.6951	0.4460
No log	4.375	70	0.4452	0.7223	0.4457
No log	4.5	72	0.4717	0.7461	0.4723
No log	4.625	74	0.4611	0.7342	0.4616
No log	4.75	76	0.4659	0.7176	0.4664
No log	4.875	78	0.4599	0.7400	0.4604
No log	5.0	80	0.4602	0.7421	0.4608
No log	5.125	82	0.4454	0.7202	0.4458
No log	5.25	84	0.4601	0.6651	0.4603
No log	5.375	86	0.4520	0.6947	0.4522
No log	5.5	88	0.4471	0.7201	0.4474
No log	5.625	90	0.4466	0.6992	0.4469
No log	5.75	92	0.4632	0.6619	0.4633
No log	5.875	94	0.4736	0.6395	0.4737
No log	6.0	96	0.4485	0.6918	0.4488
No log	6.125	98	0.4448	0.7136	0.4453
No log	6.25	100	0.4413	0.6987	0.4417
No log	6.375	102	0.4402	0.7137	0.4406
No log	6.5	104	0.4399	0.7159	0.4404
No log	6.625	106	0.4462	0.7259	0.4467
No log	6.75	108	0.4442	0.7166	0.4447
No log	6.875	110	0.4566	0.6877	0.4570
No log	7.0	112	0.4729	0.6802	0.4732
No log	7.125	114	0.4857	0.6695	0.4859
No log	7.25	116	0.4685	0.7000	0.4689
No log	7.375	118	0.4610	0.7218	0.4616
No log	7.5	120	0.4550	0.7174	0.4555
No log	7.625	122	0.4572	0.6892	0.4576
No log	7.75	124	0.4506	0.7015	0.4510
No log	7.875	126	0.4454	0.7045	0.4459
No log	8.0	128	0.4457	0.7284	0.4464
No log	8.125	130	0.4455	0.7283	0.4461
No log	8.25	132	0.4411	0.7309	0.4416
No log	8.375	134	0.4413	0.7252	0.4418
No log	8.5	136	0.4455	0.6852	0.4459
No log	8.625	138	0.4564	0.6632	0.4567
No log	8.75	140	0.4538	0.6632	0.4541
No log	8.875	142	0.4468	0.7039	0.4473
No log	9.0	144	0.4474	0.7177	0.4479
No log	9.125	146	0.4490	0.7156	0.4496
No log	9.25	148	0.4504	0.7156	0.4510
No log	9.375	150	0.4516	0.7127	0.4521
No log	9.5	152	0.4547	0.6898	0.4551
No log	9.625	154	0.4599	0.6702	0.4603
No log	9.75	156	0.4646	0.6602	0.4649
No log	9.875	158	0.4655	0.6602	0.4658
No log	10.0	160	0.4649	0.6602	0.4652

Framework versions

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