arabert_cross_organization_task1_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: 0.9947
• Qwk: 0.0679
• Mse: 0.9916

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	5.2049	-0.0008	5.2020
No log	0.25	4	1.9284	-0.0122	1.9256
No log	0.375	6	1.0209	0.0513	1.0167
No log	0.5	8	0.8405	0.0824	0.8373
No log	0.625	10	0.8656	0.1060	0.8628
No log	0.75	12	0.8332	0.1240	0.8306
No log	0.875	14	0.8775	0.0730	0.8754
No log	1.0	16	0.9183	0.0104	0.9162
No log	1.125	18	0.9057	0.0508	0.9032
No log	1.25	20	0.8886	0.1273	0.8859
No log	1.375	22	0.9957	0.0849	0.9930
No log	1.5	24	1.0595	0.1181	1.0564
No log	1.625	26	1.2289	0.0182	1.2258
No log	1.75	28	1.2976	0.0182	1.2948
No log	1.875	30	0.9648	0.1402	0.9617
No log	2.0	32	0.9714	0.1016	0.9684
No log	2.125	34	0.9511	0.0710	0.9483
No log	2.25	36	0.8591	0.1425	0.8564
No log	2.375	38	0.8696	0.1182	0.8667
No log	2.5	40	1.0662	0.0360	1.0635
No log	2.625	42	1.1724	0.0360	1.1696
No log	2.75	44	1.3100	0.0182	1.3071
No log	2.875	46	1.3304	0.0182	1.3275
No log	3.0	48	1.0676	0.0424	1.0645
No log	3.125	50	0.9732	0.0668	0.9701
No log	3.25	52	1.1173	0.0279	1.1143
No log	3.375	54	1.2420	0.0182	1.2393
No log	3.5	56	1.1410	0.0155	1.1382
No log	3.625	58	0.9316	0.0268	0.9285
No log	3.75	60	0.8907	0.1122	0.8876
No log	3.875	62	1.0183	0.0253	1.0153
No log	4.0	64	1.1271	0.0279	1.1242
No log	4.125	66	1.1742	0.0300	1.1712
No log	4.25	68	1.2066	0.0682	1.2034
No log	4.375	70	1.2604	0.0377	1.2572
No log	4.5	72	1.1679	0.0830	1.1646
No log	4.625	74	1.1770	0.0966	1.1739
No log	4.75	76	1.1163	0.0966	1.1131
No log	4.875	78	0.9754	0.0695	0.9721
No log	5.0	80	0.9489	0.0767	0.9456
No log	5.125	82	0.9900	0.0994	0.9868
No log	5.25	84	0.8622	0.0654	0.8588
No log	5.375	86	0.8621	0.1028	0.8586
No log	5.5	88	1.0043	0.0807	1.0011
No log	5.625	90	1.0565	0.0448	1.0533
No log	5.75	92	0.9899	0.0848	0.9866
No log	5.875	94	1.1141	0.0466	1.1111
No log	6.0	96	1.3040	0.0906	1.3012
No log	6.125	98	1.2856	0.1112	1.2829
No log	6.25	100	1.3671	0.0962	1.3644
No log	6.375	102	1.2601	0.1091	1.2574
No log	6.5	104	1.2039	0.1595	1.2011
No log	6.625	106	1.1272	0.0913	1.1244
No log	6.75	108	1.0754	0.0958	1.0725
No log	6.875	110	1.0818	0.0777	1.0790
No log	7.0	112	1.0175	0.0670	1.0146
No log	7.125	114	0.9552	0.0569	0.9521
No log	7.25	116	0.8938	0.1278	0.8906
No log	7.375	118	0.9486	0.0697	0.9455
No log	7.5	120	0.9351	0.0773	0.9319
No log	7.625	122	0.8928	0.0870	0.8895
No log	7.75	124	0.8558	0.1373	0.8524
No log	7.875	126	0.8561	0.1606	0.8527
No log	8.0	128	0.9205	0.0389	0.9174
No log	8.125	130	1.0514	0.0941	1.0484
No log	8.25	132	1.0795	0.1246	1.0765
No log	8.375	134	1.0151	0.0977	1.0120
No log	8.5	136	0.9815	0.0716	0.9784
No log	8.625	138	0.9817	0.0668	0.9786
No log	8.75	140	0.9721	0.0597	0.9690
No log	8.875	142	0.9865	0.0668	0.9834
No log	9.0	144	0.9956	0.0716	0.9925
No log	9.125	146	0.9824	0.0807	0.9793
No log	9.25	148	0.9599	0.0721	0.9568
No log	9.375	150	0.9488	0.0858	0.9456
No log	9.5	152	0.9443	0.0858	0.9411
No log	9.625	154	0.9603	0.0721	0.9572
No log	9.75	156	0.9767	0.0770	0.9735
No log	9.875	158	0.9906	0.0679	0.9874
No log	10.0	160	0.9947	0.0679	0.9916

Framework versions

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