arabert_cross_organization_task4_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: 1.7693
• Qwk: -0.0357
• Mse: 1.7659

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	0.1111	2	4.1886	0.0011	4.1850
No log	0.2222	4	2.2640	0.0	2.2614
No log	0.3333	6	1.3362	0.0183	1.3332
No log	0.4444	8	0.9155	0.0	0.9122
No log	0.5556	10	0.9188	-0.0048	0.9152
No log	0.6667	12	0.8902	-0.0315	0.8872
No log	0.7778	14	0.8950	0.0741	0.8923
No log	0.8889	16	0.8770	0.0875	0.8744
No log	1.0	18	0.8807	0.1030	0.8782
No log	1.1111	20	0.8915	-0.0099	0.8887
No log	1.2222	22	0.9457	0.0424	0.9431
No log	1.3333	24	1.0318	0.0182	1.0292
No log	1.4444	26	1.1586	0.0	1.1562
No log	1.5556	28	1.2936	0.0	1.2909
No log	1.6667	30	1.4219	0.0	1.4193
No log	1.7778	32	1.4842	0.0	1.4816
No log	1.8889	34	1.4765	0.0	1.4737
No log	2.0	36	1.4797	0.0	1.4769
No log	2.1111	38	1.4454	0.0	1.4425
No log	2.2222	40	1.5693	0.0	1.5665
No log	2.3333	42	1.6763	0.0	1.6737
No log	2.4444	44	1.6705	0.0	1.6677
No log	2.5556	46	1.4309	0.0182	1.4278
No log	2.6667	48	1.3201	0.0536	1.3169
No log	2.7778	50	1.5069	0.0182	1.5037
No log	2.8889	52	1.6390	0.0	1.6360
No log	3.0	54	1.5712	0.0	1.5682
No log	3.1111	56	1.4444	0.0182	1.4413
No log	3.2222	58	1.3827	0.0182	1.3796
No log	3.3333	60	1.5276	0.0	1.5245
No log	3.4444	62	1.5680	0.0025	1.5650
No log	3.5556	64	1.5963	0.0328	1.5931
No log	3.6667	66	1.5027	-0.0118	1.4994
No log	3.7778	68	1.5906	0.0485	1.5873
No log	3.8889	70	1.7316	0.0225	1.7284
No log	4.0	72	1.6752	0.0160	1.6719
No log	4.1111	74	1.4644	0.0136	1.4611
No log	4.2222	76	1.2946	0.0498	1.2914
No log	4.3333	78	1.3892	0.0571	1.3860
No log	4.4444	80	1.6935	0.0183	1.6903
No log	4.5556	82	1.8327	0.0026	1.8295
No log	4.6667	84	1.7518	0.0272	1.7486
No log	4.7778	86	1.5355	0.0437	1.5322
No log	4.8889	88	1.4546	0.0750	1.4513
No log	5.0	90	1.6165	0.0875	1.6132
No log	5.1111	92	1.8925	0.0054	1.8891
No log	5.2222	94	1.9400	-0.0784	1.9366
No log	5.3333	96	1.7667	0.0162	1.7634
No log	5.4444	98	1.5714	0.0700	1.5681
No log	5.5556	100	1.6393	0.0586	1.6360
No log	5.6667	102	1.7094	0.0144	1.7060
No log	5.7778	104	1.8384	0.0162	1.8349
No log	5.8889	106	1.7972	0.0130	1.7937
No log	6.0	108	1.7520	0.0113	1.7486
No log	6.1111	110	1.6669	0.0264	1.6636
No log	6.2222	112	1.6880	0.0301	1.6846
No log	6.3333	114	1.7562	0.0096	1.7528
No log	6.4444	116	1.7980	0.0228	1.7947
No log	6.5556	118	1.8220	-0.0199	1.8187
No log	6.6667	120	1.7461	0.0746	1.7429
No log	6.7778	122	1.5563	0.0272	1.5532
No log	6.8889	124	1.4185	0.0655	1.4154
No log	7.0	126	1.4690	0.0697	1.4658
No log	7.1111	128	1.6412	0.0722	1.6381
No log	7.2222	130	1.8101	-0.0325	1.8069
No log	7.3333	132	1.8161	-0.0588	1.8129
No log	7.4444	134	1.6991	0.0758	1.6959
No log	7.5556	136	1.5693	0.0476	1.5661
No log	7.6667	138	1.5215	0.0437	1.5183
No log	7.7778	140	1.5990	0.0699	1.5957
No log	7.8889	142	1.7382	0.0363	1.7348
No log	8.0	144	1.8943	-0.0587	1.8909
No log	8.1111	146	1.9468	-0.0675	1.9434
No log	8.2222	148	1.9074	-0.0587	1.9040
No log	8.3333	150	1.8074	-0.0208	1.8040
No log	8.4444	152	1.7220	0.0241	1.7186
No log	8.5556	154	1.6877	0.0352	1.6843
No log	8.6667	156	1.7113	0.0222	1.7079
No log	8.7778	158	1.7927	-0.0208	1.7893
No log	8.8889	160	1.8848	-0.0477	1.8813
No log	9.0	162	1.9430	-0.0613	1.9396
No log	9.1111	164	1.9388	-0.0613	1.9354
No log	9.2222	166	1.8894	-0.0477	1.8860
No log	9.3333	168	1.8314	-0.0424	1.8279
No log	9.4444	170	1.7625	-0.0357	1.7591
No log	9.5556	172	1.7335	0.0264	1.7302
No log	9.6667	174	1.7407	0.0264	1.7373
No log	9.7778	176	1.7576	-0.0102	1.7543
No log	9.8889	178	1.7644	-0.0357	1.7610
No log	10.0	180	1.7693	-0.0357	1.7659

Framework versions

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