arabert_cross_organization_task3_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: 0.6478
• Qwk: 0.4607
• Mse: 0.6478

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	3.2900	0.0231	3.2900
No log	0.25	4	1.3894	0.0822	1.3894
No log	0.375	6	0.7800	0.1877	0.7800
No log	0.5	8	1.0858	0.2133	1.0858
No log	0.625	10	1.0129	0.2378	1.0129
No log	0.75	12	0.8148	0.3122	0.8148
No log	0.875	14	0.8640	0.3294	0.8640
No log	1.0	16	0.5808	0.5261	0.5808
No log	1.125	18	0.5809	0.5472	0.5809
No log	1.25	20	0.6891	0.4535	0.6891
No log	1.375	22	0.8095	0.4026	0.8095
No log	1.5	24	0.6001	0.4892	0.6001
No log	1.625	26	0.5488	0.5204	0.5488
No log	1.75	28	0.6546	0.4189	0.6546
No log	1.875	30	0.5905	0.4711	0.5905
No log	2.0	32	0.5350	0.5446	0.5350
No log	2.125	34	0.6019	0.4942	0.6019
No log	2.25	36	0.6144	0.4795	0.6144
No log	2.375	38	0.5350	0.5333	0.5350
No log	2.5	40	0.5122	0.5472	0.5122
No log	2.625	42	0.5046	0.5160	0.5046
No log	2.75	44	0.5608	0.4614	0.5608
No log	2.875	46	0.6094	0.4437	0.6094
No log	3.0	48	0.5336	0.5068	0.5336
No log	3.125	50	0.5450	0.5314	0.5450
No log	3.25	52	0.5818	0.5196	0.5818
No log	3.375	54	0.6351	0.4649	0.6351
No log	3.5	56	0.6078	0.4611	0.6078
No log	3.625	58	0.5257	0.4924	0.5257
No log	3.75	60	0.5373	0.4903	0.5373
No log	3.875	62	0.6198	0.4443	0.6198
No log	4.0	64	0.6554	0.4271	0.6554
No log	4.125	66	0.5949	0.4804	0.5949
No log	4.25	68	0.5758	0.5021	0.5758
No log	4.375	70	0.6050	0.4669	0.6050
No log	4.5	72	0.6617	0.4443	0.6617
No log	4.625	74	0.5939	0.4755	0.5939
No log	4.75	76	0.5429	0.5333	0.5429
No log	4.875	78	0.5670	0.4951	0.5670
No log	5.0	80	0.6057	0.4790	0.6057
No log	5.125	82	0.5475	0.5271	0.5475
No log	5.25	84	0.5293	0.5556	0.5293
No log	5.375	86	0.5920	0.4961	0.5920
No log	5.5	88	0.8172	0.3957	0.8172
No log	5.625	90	0.8892	0.3757	0.8892
No log	5.75	92	0.7109	0.4460	0.7109
No log	5.875	94	0.5766	0.5031	0.5766
No log	6.0	96	0.5308	0.5421	0.5308
No log	6.125	98	0.5523	0.5081	0.5523
No log	6.25	100	0.6890	0.4239	0.6890
No log	6.375	102	0.7810	0.4232	0.7810
No log	6.5	104	0.6877	0.4432	0.6877
No log	6.625	106	0.5825	0.4691	0.5825
No log	6.75	108	0.5649	0.4830	0.5649
No log	6.875	110	0.5645	0.4955	0.5645
No log	7.0	112	0.6053	0.4622	0.6053
No log	7.125	114	0.6870	0.4274	0.6870
No log	7.25	116	0.7007	0.4215	0.7007
No log	7.375	118	0.6387	0.4483	0.6387
No log	7.5	120	0.5709	0.5018	0.5709
No log	7.625	122	0.5594	0.5102	0.5594
No log	7.75	124	0.6003	0.4811	0.6003
No log	7.875	126	0.6909	0.4242	0.6909
No log	8.0	128	0.7230	0.4165	0.7230
No log	8.125	130	0.6705	0.4395	0.6705
No log	8.25	132	0.6041	0.4682	0.6041
No log	8.375	134	0.5802	0.4903	0.5802
No log	8.5	136	0.5853	0.4903	0.5853
No log	8.625	138	0.6211	0.4634	0.6211
No log	8.75	140	0.6644	0.4483	0.6644
No log	8.875	142	0.6824	0.4514	0.6824
No log	9.0	144	0.6713	0.4472	0.6713
No log	9.125	146	0.6586	0.4498	0.6586
No log	9.25	148	0.6406	0.4592	0.6406
No log	9.375	150	0.6229	0.4711	0.6229
No log	9.5	152	0.6220	0.4720	0.6220
No log	9.625	154	0.6319	0.4686	0.6319
No log	9.75	156	0.6419	0.4623	0.6419
No log	9.875	158	0.6451	0.4623	0.6451
No log	10.0	160	0.6478	0.4607	0.6478

Framework versions

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