arabert_cross_organization_task6_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: 1.0752
• Qwk: 0.1211
• Mse: 1.0727

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	3.7816	0.0007	3.7767
No log	0.2353	4	1.3542	0.0008	1.3493
No log	0.3529	6	0.8381	0.0381	0.8355
No log	0.4706	8	0.8831	-0.0815	0.8796
No log	0.5882	10	0.8969	0.0429	0.8931
No log	0.7059	12	0.8250	0.0520	0.8218
No log	0.8235	14	1.0019	0.0182	0.9999
No log	0.9412	16	1.1717	-0.0025	1.1701
No log	1.0588	18	1.2136	-0.0025	1.2120
No log	1.1765	20	1.1606	0.0	1.1587
No log	1.2941	22	1.2446	0.0	1.2426
No log	1.4118	24	1.3967	0.0	1.3948
No log	1.5294	26	1.4847	0.0253	1.4827
No log	1.6471	28	1.3244	0.0231	1.3221
No log	1.7647	30	1.0785	0.0360	1.0760
No log	1.8824	32	0.8498	0.1972	0.8470
No log	2.0	34	0.9134	0.1247	0.9107
No log	2.1176	36	1.1156	0.0182	1.1130
No log	2.2353	38	1.0973	0.0	1.0948
No log	2.3529	40	1.0822	0.0182	1.0794
No log	2.4706	42	1.0078	0.0789	1.0047
No log	2.5882	44	1.1175	0.0880	1.1144
No log	2.7059	46	1.3284	0.0465	1.3253
No log	2.8235	48	1.6008	0.0931	1.5976
No log	2.9412	50	2.1079	0.0276	2.1049
No log	3.0588	52	2.2734	0.0412	2.2705
No log	3.1765	54	1.8347	0.1369	1.8320
No log	3.2941	56	1.3122	0.0884	1.3096
No log	3.4118	58	1.2595	0.0664	1.2569
No log	3.5294	60	1.4282	0.1006	1.4256
No log	3.6471	62	1.5462	0.1413	1.5436
No log	3.7647	64	1.3829	0.0593	1.3803
No log	3.8824	66	0.9939	0.1230	0.9914
No log	4.0	68	0.8054	0.1788	0.8030
No log	4.1176	70	0.7868	0.1816	0.7845
No log	4.2353	72	0.9015	0.1607	0.8991
No log	4.3529	74	1.2405	0.0483	1.2379
No log	4.4706	76	1.4275	0.0040	1.4247
No log	4.5882	78	1.3885	0.0201	1.3856
No log	4.7059	80	1.2664	0.0431	1.2634
No log	4.8235	82	1.0916	0.0750	1.0885
No log	4.9412	84	1.0885	0.0159	1.0854
No log	5.0588	86	1.1736	0.0869	1.1705
No log	5.1765	88	1.3914	0.1105	1.3885
No log	5.2941	90	1.5037	0.0839	1.5010
No log	5.4118	92	1.4052	0.0391	1.4026
No log	5.5294	94	1.2441	0.0662	1.2417
No log	5.6471	96	1.1952	0.0646	1.1929
No log	5.7647	98	1.1442	0.0814	1.1419
No log	5.8824	100	1.1950	0.0855	1.1926
No log	6.0	102	1.3191	0.0880	1.3167
No log	6.1176	104	1.4094	0.0999	1.4070
No log	6.2353	106	1.2755	0.1507	1.2730
No log	6.3529	108	1.0797	0.1262	1.0770
No log	6.4706	110	0.9830	0.1381	0.9804
No log	6.5882	112	0.9647	0.1405	0.9620
No log	6.7059	114	1.0370	0.1146	1.0345
No log	6.8235	116	1.1227	0.0885	1.1203
No log	6.9412	118	1.2057	0.1026	1.2033
No log	7.0588	120	1.2261	0.0518	1.2238
No log	7.1765	122	1.2047	0.0652	1.2023
No log	7.2941	124	1.0652	0.1195	1.0628
No log	7.4118	126	0.9603	0.1804	0.9578
No log	7.5294	128	0.9421	0.1812	0.9395
No log	7.6471	130	1.0068	0.1111	1.0041
No log	7.7647	132	1.1676	0.1367	1.1650
No log	7.8824	134	1.3081	0.1382	1.3055
No log	8.0	136	1.3954	0.1123	1.3929
No log	8.1176	138	1.3854	0.1123	1.3829
No log	8.2353	140	1.2987	0.1203	1.2962
No log	8.3529	142	1.1429	0.1190	1.1405
No log	8.4706	144	1.0515	0.1232	1.0491
No log	8.5882	146	1.0416	0.1371	1.0392
No log	8.7059	148	1.0625	0.1438	1.0601
No log	8.8235	150	1.0904	0.1152	1.0880
No log	8.9412	152	1.1358	0.1190	1.1334
No log	9.0588	154	1.1786	0.1043	1.1762
No log	9.1765	156	1.1978	0.1043	1.1954
No log	9.2941	158	1.1835	0.1043	1.1810
No log	9.4118	160	1.1423	0.1190	1.1399
No log	9.5294	162	1.1135	0.1336	1.1110
No log	9.6471	164	1.1006	0.1132	1.0981
No log	9.7647	166	1.0926	0.1277	1.0902
No log	9.8824	168	1.0805	0.1211	1.0780
No log	10.0	170	1.0752	0.1211	1.0727

Framework versions

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