arabert_cross_organization_task3_fold3

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: 0.5370
• Qwk: 0.7944
• Mse: 0.5370

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	1.7580	0.0890	1.7580
No log	0.25	4	1.2272	0.4035	1.2272
No log	0.375	6	0.9464	0.5403	0.9464
No log	0.5	8	0.7613	0.6565	0.7613
No log	0.625	10	0.6541	0.7029	0.6541
No log	0.75	12	0.6490	0.7391	0.6490
No log	0.875	14	0.7327	0.7661	0.7327
No log	1.0	16	0.5927	0.6919	0.5927
No log	1.125	18	0.5964	0.6599	0.5964
No log	1.25	20	0.6354	0.7806	0.6354
No log	1.375	22	0.8402	0.7721	0.8402
No log	1.5	24	0.6567	0.7904	0.6567
No log	1.625	26	0.5279	0.7541	0.5279
No log	1.75	28	0.5231	0.7390	0.5231
No log	1.875	30	0.5289	0.7813	0.5289
No log	2.0	32	0.5925	0.7797	0.5925
No log	2.125	34	0.5332	0.7806	0.5332
No log	2.25	36	0.5328	0.7818	0.5328
No log	2.375	38	0.5917	0.7871	0.5917
No log	2.5	40	0.5461	0.7940	0.5461
No log	2.625	42	0.5813	0.7869	0.5813
No log	2.75	44	0.5361	0.7835	0.5361
No log	2.875	46	0.4895	0.7367	0.4895
No log	3.0	48	0.4986	0.7699	0.4986
No log	3.125	50	0.5492	0.7845	0.5492
No log	3.25	52	0.6585	0.7981	0.6585
No log	3.375	54	0.5821	0.7973	0.5821
No log	3.5	56	0.4999	0.7692	0.4999
No log	3.625	58	0.5101	0.7700	0.5101
No log	3.75	60	0.5803	0.7877	0.5803
No log	3.875	62	0.5825	0.78	0.5825
No log	4.0	64	0.5152	0.7826	0.5152
No log	4.125	66	0.5153	0.7737	0.5153
No log	4.25	68	0.5298	0.7824	0.5298
No log	4.375	70	0.4955	0.7601	0.4955
No log	4.5	72	0.5345	0.7834	0.5345
No log	4.625	74	0.5955	0.7877	0.5955
No log	4.75	76	0.5233	0.7839	0.5233
No log	4.875	78	0.4939	0.7760	0.4939
No log	5.0	80	0.4893	0.7615	0.4893
No log	5.125	82	0.5115	0.7835	0.5115
No log	5.25	84	0.5169	0.7812	0.5169
No log	5.375	86	0.5423	0.7856	0.5423
No log	5.5	88	0.5269	0.7741	0.5269
No log	5.625	90	0.5073	0.7634	0.5073
No log	5.75	92	0.5209	0.7721	0.5209
No log	5.875	94	0.6436	0.8138	0.6436
No log	6.0	96	0.7068	0.8088	0.7068
No log	6.125	98	0.6021	0.7992	0.6021
No log	6.25	100	0.5022	0.7464	0.5022
No log	6.375	102	0.4976	0.7471	0.4976
No log	6.5	104	0.5144	0.7552	0.5144
No log	6.625	106	0.5770	0.7931	0.5770
No log	6.75	108	0.5866	0.7931	0.5866
No log	6.875	110	0.5239	0.7649	0.5239
No log	7.0	112	0.4930	0.7531	0.4930
No log	7.125	114	0.4884	0.7636	0.4884
No log	7.25	116	0.5129	0.7744	0.5129
No log	7.375	118	0.5771	0.7982	0.5771
No log	7.5	120	0.5703	0.7982	0.5703
No log	7.625	122	0.5147	0.7681	0.5147
No log	7.75	124	0.4973	0.7739	0.4973
No log	7.875	126	0.4888	0.7760	0.4888
No log	8.0	128	0.5035	0.7817	0.5035
No log	8.125	130	0.5124	0.7723	0.5124
No log	8.25	132	0.5107	0.7782	0.5107
No log	8.375	134	0.5195	0.7768	0.5195
No log	8.5	136	0.5271	0.7724	0.5271
No log	8.625	138	0.5365	0.7820	0.5365
No log	8.75	140	0.5329	0.7862	0.5329
No log	8.875	142	0.5215	0.7883	0.5215
No log	9.0	144	0.5191	0.7827	0.5191
No log	9.125	146	0.5120	0.7830	0.5120
No log	9.25	148	0.5044	0.7814	0.5044
No log	9.375	150	0.5086	0.7870	0.5086
No log	9.5	152	0.5195	0.7956	0.5195
No log	9.625	154	0.5294	0.7955	0.5294
No log	9.75	156	0.5377	0.7944	0.5377
No log	9.875	158	0.5380	0.7958	0.5380
No log	10.0	160	0.5370	0.7944	0.5370

Framework versions

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