arabert_cross_organization_task7_fold0

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: 0.7058
• Qwk: 0.5975
• Mse: 0.7057

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	0.1333	2	3.8864	0.0541	3.8830
No log	0.2667	4	2.5481	0.0079	2.5460
No log	0.4	6	1.6996	0.2037	1.6983
No log	0.5333	8	1.3776	0.2939	1.3765
No log	0.6667	10	1.3355	0.3347	1.3345
No log	0.8	12	1.2773	0.3590	1.2767
No log	0.9333	14	1.4284	0.3342	1.4277
No log	1.0667	16	1.1577	0.3897	1.1573
No log	1.2	18	0.9645	0.4536	0.9641
No log	1.3333	20	1.1001	0.4324	1.0996
No log	1.4667	22	1.0995	0.4399	1.0990
No log	1.6	24	0.9787	0.4704	0.9783
No log	1.7333	26	0.8116	0.5438	0.8112
No log	1.8667	28	0.8107	0.5471	0.8103
No log	2.0	30	0.9069	0.4983	0.9063
No log	2.1333	32	0.8730	0.5346	0.8726
No log	2.2667	34	0.9141	0.5284	0.9138
No log	2.4	36	0.7576	0.6068	0.7578
No log	2.5333	38	0.7560	0.6087	0.7563
No log	2.6667	40	0.8005	0.6059	0.8007
No log	2.8	42	0.9430	0.5399	0.9430
No log	2.9333	44	0.8990	0.5542	0.8990
No log	3.0667	46	0.8494	0.5672	0.8494
No log	3.2	48	0.8095	0.5752	0.8094
No log	3.3333	50	0.7424	0.6012	0.7425
No log	3.4667	52	0.7009	0.6100	0.7011
No log	3.6	54	0.7078	0.6055	0.7080
No log	3.7333	56	0.8222	0.5652	0.8223
No log	3.8667	58	0.8211	0.5742	0.8212
No log	4.0	60	0.7561	0.5870	0.7562
No log	4.1333	62	0.6981	0.6055	0.6982
No log	4.2667	64	0.6970	0.6051	0.6971
No log	4.4	66	0.7327	0.5991	0.7326
No log	4.5333	68	0.8809	0.5424	0.8806
No log	4.6667	70	0.8266	0.5559	0.8263
No log	4.8	72	0.7118	0.5954	0.7117
No log	4.9333	74	0.6923	0.6040	0.6923
No log	5.0667	76	0.7259	0.5893	0.7259
No log	5.2	78	0.7425	0.5832	0.7425
No log	5.3333	80	0.6843	0.6051	0.6843
No log	5.4667	82	0.6647	0.6118	0.6647
No log	5.6	84	0.6805	0.6104	0.6804
No log	5.7333	86	0.7021	0.5992	0.7019
No log	5.8667	88	0.6896	0.6158	0.6894
No log	6.0	90	0.6658	0.6204	0.6657
No log	6.1333	92	0.6774	0.6118	0.6774
No log	6.2667	94	0.7472	0.5748	0.7471
No log	6.4	96	0.8444	0.5491	0.8442
No log	6.5333	98	0.8227	0.5502	0.8225
No log	6.6667	100	0.7455	0.5671	0.7453
No log	6.8	102	0.7062	0.5907	0.7060
No log	6.9333	104	0.6458	0.6143	0.6457
No log	7.0667	106	0.6289	0.6250	0.6288
No log	7.2	108	0.6373	0.6205	0.6371
No log	7.3333	110	0.6669	0.6113	0.6666
No log	7.4667	112	0.7012	0.6035	0.7009
No log	7.6	114	0.6942	0.5999	0.6939
No log	7.7333	116	0.6669	0.6083	0.6668
No log	7.8667	118	0.6470	0.6177	0.6469
No log	8.0	120	0.6439	0.6292	0.6438
No log	8.1333	122	0.6501	0.6098	0.6500
No log	8.2667	124	0.6801	0.6013	0.6800
No log	8.4	126	0.7382	0.5951	0.7380
No log	8.5333	128	0.7558	0.5885	0.7556
No log	8.6667	130	0.7227	0.5947	0.7225
No log	8.8	132	0.6780	0.6014	0.6780
No log	8.9333	134	0.6554	0.6101	0.6554
No log	9.0667	136	0.6520	0.6154	0.6520
No log	9.2	138	0.6597	0.6100	0.6596
No log	9.3333	140	0.6690	0.6141	0.6689
No log	9.4667	142	0.6790	0.6041	0.6789
No log	9.6	144	0.6910	0.5976	0.6909
No log	9.7333	146	0.7025	0.5975	0.7024
No log	9.8667	148	0.7059	0.5975	0.7057
No log	10.0	150	0.7058	0.5975	0.7057

Framework versions

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