arabert_cross_organization_task2_fold2

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: 1.0548
• Qwk: 0.1458
• Mse: 1.0548

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	0.1176	2	4.4485	-0.0131	4.4485
No log	0.2353	4	1.8009	-0.0153	1.8009
No log	0.3529	6	1.1944	0.0437	1.1944
No log	0.4706	8	1.2459	-0.0439	1.2459
No log	0.5882	10	1.1688	-0.0450	1.1688
No log	0.7059	12	1.1521	0.0379	1.1521
No log	0.8235	14	1.1603	-0.0096	1.1603
No log	0.9412	16	1.1948	-0.0723	1.1948
No log	1.0588	18	1.3189	-0.0617	1.3189
No log	1.1765	20	1.4608	-0.0017	1.4608
No log	1.2941	22	1.2409	-0.0661	1.2409
No log	1.4118	24	1.1260	0.0513	1.1260
No log	1.5294	26	1.1582	0.0639	1.1582
No log	1.6471	28	1.1974	-0.0637	1.1974
No log	1.7647	30	1.2427	-0.0072	1.2427
No log	1.8824	32	1.2670	-0.0237	1.2670
No log	2.0	34	1.2862	-0.0775	1.2862
No log	2.1176	36	1.2811	-0.0301	1.2811
No log	2.2353	38	1.2493	0.0198	1.2493
No log	2.3529	40	1.3023	-0.0524	1.3023
No log	2.4706	42	1.1620	0.0963	1.1620
No log	2.5882	44	1.2006	0.0203	1.2006
No log	2.7059	46	1.1721	-0.0203	1.1721
No log	2.8235	48	1.1251	0.0909	1.1251
No log	2.9412	50	1.1646	0.0614	1.1646
No log	3.0588	52	1.1537	0.1100	1.1537
No log	3.1765	54	1.1619	0.0397	1.1619
No log	3.2941	56	1.1446	0.0356	1.1446
No log	3.4118	58	1.1100	0.1366	1.1100
No log	3.5294	60	1.0950	0.0927	1.0950
No log	3.6471	62	1.1914	0.0668	1.1914
No log	3.7647	64	1.1642	0.0734	1.1642
No log	3.8824	66	1.0702	0.1452	1.0702
No log	4.0	68	1.0704	0.1474	1.0704
No log	4.1176	70	1.1170	0.0391	1.1170
No log	4.2353	72	1.1800	0.0595	1.1800
No log	4.3529	74	1.0993	0.1283	1.0993
No log	4.4706	76	1.1934	-0.0135	1.1934
No log	4.5882	78	1.3754	-0.0259	1.3754
No log	4.7059	80	1.2204	-0.0135	1.2204
No log	4.8235	82	1.1114	0.1016	1.1114
No log	4.9412	84	1.3701	0.0289	1.3701
No log	5.0588	86	1.3316	0.0728	1.3316
No log	5.1765	88	1.0952	0.1185	1.0952
No log	5.2941	90	1.1314	0.0453	1.1314
No log	5.4118	92	1.2064	-0.0035	1.2064
No log	5.5294	94	1.1624	0.0524	1.1624
No log	5.6471	96	1.0616	0.1805	1.0616
No log	5.7647	98	1.0707	0.1579	1.0707
No log	5.8824	100	1.0682	0.1543	1.0682
No log	6.0	102	1.0755	0.1611	1.0755
No log	6.1176	104	1.1047	0.1146	1.1047
No log	6.2353	106	1.0839	0.1630	1.0839
No log	6.3529	108	1.1070	0.1144	1.1070
No log	6.4706	110	1.1227	0.1144	1.1227
No log	6.5882	112	1.1078	0.1050	1.1078
No log	6.7059	114	1.1016	0.1871	1.1016
No log	6.8235	116	1.0973	0.1835	1.0973
No log	6.9412	118	1.0879	0.1622	1.0879
No log	7.0588	120	1.1030	0.1077	1.1030
No log	7.1765	122	1.0923	0.1497	1.0923
No log	7.2941	124	1.0894	0.1520	1.0894
No log	7.4118	126	1.0755	0.1253	1.0755
No log	7.5294	128	1.0677	0.1253	1.0677
No log	7.6471	130	1.0610	0.1520	1.0610
No log	7.7647	132	1.1106	0.0794	1.1106
No log	7.8824	134	1.0932	0.0888	1.0932
No log	8.0	136	1.0392	0.1335	1.0392
No log	8.1176	138	1.0756	0.1295	1.0756
No log	8.2353	140	1.1032	0.1371	1.1032
No log	8.3529	142	1.0707	0.1344	1.0707
No log	8.4706	144	1.0399	0.1754	1.0399
No log	8.5882	146	1.0486	0.1205	1.0486
No log	8.7059	148	1.0606	0.1379	1.0606
No log	8.8235	150	1.0717	0.1241	1.0717
No log	8.9412	152	1.0674	0.1379	1.0674
No log	9.0588	154	1.0527	0.1066	1.0527
No log	9.1765	156	1.0449	0.1458	1.0449
No log	9.2941	158	1.0540	0.1907	1.0540
No log	9.4118	160	1.0626	0.1548	1.0626
No log	9.5294	162	1.0605	0.1577	1.0605
No log	9.6471	164	1.0583	0.1605	1.0583
No log	9.7647	166	1.0560	0.1458	1.0560
No log	9.8824	168	1.0549	0.1458	1.0549
No log	10.0	170	1.0548	0.1458	1.0548

Framework versions

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