arabert_cross_organization_task2_fold1

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: 0.9665
• Qwk: 0.1843
• Mse: 0.9638

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	4.9621	0.0014	4.9593
No log	0.25	4	2.1315	-0.0199	2.1286
No log	0.375	6	0.9548	0.0027	0.9513
No log	0.5	8	0.8530	0.0678	0.8507
No log	0.625	10	0.7928	0.1306	0.7900
No log	0.75	12	0.7710	0.1552	0.7682
No log	0.875	14	0.8786	-0.0670	0.8766
No log	1.0	16	0.9254	-0.0565	0.9242
No log	1.125	18	1.0088	-0.0232	1.0068
No log	1.25	20	0.8151	0.1520	0.8123
No log	1.375	22	0.7943	0.0808	0.7913
No log	1.5	24	0.8986	0.0767	0.8958
No log	1.625	26	0.9046	0.0935	0.9018
No log	1.75	28	0.8005	0.0962	0.7975
No log	1.875	30	0.8246	0.1261	0.8217
No log	2.0	32	1.0455	0.0565	1.0427
No log	2.125	34	1.0848	0.0536	1.0819
No log	2.25	36	0.8432	0.1284	0.8401
No log	2.375	38	0.8148	0.1252	0.8116
No log	2.5	40	0.9950	0.0565	0.9919
No log	2.625	42	1.3659	0.0	1.3632
No log	2.75	44	1.3473	0.0	1.3447
No log	2.875	46	1.0770	0.0536	1.0742
No log	3.0	48	0.9980	0.1016	0.9952
No log	3.125	50	1.2032	0.0	1.2005
No log	3.25	52	1.5241	0.0	1.5217
No log	3.375	54	1.5210	0.0231	1.5187
No log	3.5	56	1.3342	0.0	1.3319
No log	3.625	58	1.1036	0.0	1.1011
No log	3.75	60	1.0121	0.0536	1.0095
No log	3.875	62	1.1441	0.0	1.1415
No log	4.0	64	1.3639	0.0	1.3614
No log	4.125	66	1.3735	0.0025	1.3709
No log	4.25	68	1.0825	0.0	1.0799
No log	4.375	70	0.9120	0.1641	0.9093
No log	4.5	72	0.9506	0.1277	0.9480
No log	4.625	74	1.1517	0.0	1.1492
No log	4.75	76	1.2276	0.0	1.2251
No log	4.875	78	1.1183	0.0	1.1159
No log	5.0	80	1.0068	0.0333	1.0043
No log	5.125	82	1.0353	0.0360	1.0328
No log	5.25	84	1.2149	0.0025	1.2124
No log	5.375	86	1.3871	-0.0077	1.3847
No log	5.5	88	1.3159	-0.0052	1.3135
No log	5.625	90	1.1412	0.0600	1.1386
No log	5.75	92	1.1161	0.0755	1.1135
No log	5.875	94	1.0471	0.1059	1.0444
No log	6.0	96	0.9768	0.1096	0.9740
No log	6.125	98	0.8671	0.2312	0.8642
No log	6.25	100	0.8025	0.1206	0.7994
No log	6.375	102	0.8027	0.1414	0.7995
No log	6.5	104	0.8550	0.1743	0.8519
No log	6.625	106	0.9815	0.0889	0.9787
No log	6.75	108	1.0738	0.0846	1.0710
No log	6.875	110	1.1903	0.1144	1.1876
No log	7.0	112	1.2101	0.1122	1.2074
No log	7.125	114	1.2463	0.0812	1.2436
No log	7.25	116	1.3137	0.0117	1.3110
No log	7.375	118	1.3237	-0.0052	1.3210
No log	7.5	120	1.2538	0.0649	1.2512
No log	7.625	122	1.2188	0.1004	1.2162
No log	7.75	124	1.1146	0.1281	1.1120
No log	7.875	126	1.0332	0.1480	1.0306
No log	8.0	128	0.9680	0.1644	0.9653
No log	8.125	130	0.9608	0.1792	0.9581
No log	8.25	132	0.9993	0.1681	0.9966
No log	8.375	134	1.0373	0.0889	1.0347
No log	8.5	136	1.0747	0.1177	1.0721
No log	8.625	138	1.0865	0.1030	1.0839
No log	8.75	140	1.0732	0.1486	1.0707
No log	8.875	142	1.0424	0.1505	1.0398
No log	9.0	144	1.0425	0.1505	1.0399
No log	9.125	146	1.0220	0.1587	1.0194
No log	9.25	148	1.0055	0.2009	1.0029
No log	9.375	150	0.9949	0.1941	0.9923
No log	9.5	152	0.9725	0.1875	0.9698
No log	9.625	154	0.9600	0.2145	0.9574
No log	9.75	156	0.9593	0.2145	0.9566
No log	9.875	158	0.9660	0.1843	0.9634
No log	10.0	160	0.9665	0.1843	0.9638

Framework versions

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