arabert_cross_organization_task1_fold4

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: 0.4588
• Qwk: 0.6885
• Mse: 0.4588

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	3.0429	0.0044	3.0429
No log	0.25	4	1.6578	0.1373	1.6578
No log	0.375	6	0.9445	0.3393	0.9445
No log	0.5	8	0.7446	0.4700	0.7446
No log	0.625	10	0.8109	0.4239	0.8109
No log	0.75	12	0.5652	0.6068	0.5652
No log	0.875	14	0.6877	0.6129	0.6877
No log	1.0	16	0.5401	0.6022	0.5401
No log	1.125	18	0.5571	0.5613	0.5571
No log	1.25	20	0.4854	0.6440	0.4854
No log	1.375	22	0.5443	0.7366	0.5443
No log	1.5	24	0.5077	0.7444	0.5077
No log	1.625	26	0.5015	0.6266	0.5015
No log	1.75	28	0.5012	0.6164	0.5012
No log	1.875	30	0.4504	0.7043	0.4504
No log	2.0	32	0.4864	0.7187	0.4864
No log	2.125	34	0.4305	0.7243	0.4305
No log	2.25	36	0.4572	0.6579	0.4572
No log	2.375	38	0.4545	0.7032	0.4545
No log	2.5	40	0.4159	0.7123	0.4159
No log	2.625	42	0.4122	0.7591	0.4122
No log	2.75	44	0.4424	0.7617	0.4424
No log	2.875	46	0.4110	0.7600	0.4110
No log	3.0	48	0.3993	0.7372	0.3993
No log	3.125	50	0.3990	0.7391	0.3990
No log	3.25	52	0.3923	0.7306	0.3923
No log	3.375	54	0.4375	0.7685	0.4375
No log	3.5	56	0.4628	0.7698	0.4628
No log	3.625	58	0.4089	0.7365	0.4089
No log	3.75	60	0.4113	0.7238	0.4113
No log	3.875	62	0.4117	0.7308	0.4117
No log	4.0	64	0.4183	0.7175	0.4183
No log	4.125	66	0.4326	0.7175	0.4326
No log	4.25	68	0.4439	0.7360	0.4439
No log	4.375	70	0.4530	0.7375	0.4530
No log	4.5	72	0.4458	0.7040	0.4458
No log	4.625	74	0.4431	0.7054	0.4431
No log	4.75	76	0.4403	0.6980	0.4403
No log	4.875	78	0.4350	0.7144	0.4350
No log	5.0	80	0.4311	0.7511	0.4311
No log	5.125	82	0.4257	0.7418	0.4257
No log	5.25	84	0.4298	0.7174	0.4298
No log	5.375	86	0.4420	0.6877	0.4420
No log	5.5	88	0.4344	0.7174	0.4344
No log	5.625	90	0.4324	0.7146	0.4324
No log	5.75	92	0.4363	0.7566	0.4363
No log	5.875	94	0.4499	0.7689	0.4499
No log	6.0	96	0.4217	0.7367	0.4217
No log	6.125	98	0.4252	0.7237	0.4252
No log	6.25	100	0.4235	0.7141	0.4235
No log	6.375	102	0.4211	0.7230	0.4211
No log	6.5	104	0.4285	0.7493	0.4285
No log	6.625	106	0.4367	0.7530	0.4367
No log	6.75	108	0.4214	0.7457	0.4214
No log	6.875	110	0.4380	0.6930	0.4380
No log	7.0	112	0.4555	0.6727	0.4555
No log	7.125	114	0.4358	0.6947	0.4358
No log	7.25	116	0.4270	0.7277	0.4270
No log	7.375	118	0.4349	0.7457	0.4349
No log	7.5	120	0.4430	0.7382	0.4430
No log	7.625	122	0.4539	0.7257	0.4539
No log	7.75	124	0.4623	0.7204	0.4623
No log	7.875	126	0.4640	0.7110	0.4640
No log	8.0	128	0.4644	0.7115	0.4644
No log	8.125	130	0.4639	0.7095	0.4639
No log	8.25	132	0.4612	0.7073	0.4612
No log	8.375	134	0.4652	0.6865	0.4652
No log	8.5	136	0.4689	0.6753	0.4689
No log	8.625	138	0.4608	0.6849	0.4608
No log	8.75	140	0.4553	0.6907	0.4553
No log	8.875	142	0.4538	0.6930	0.4538
No log	9.0	144	0.4537	0.7172	0.4537
No log	9.125	146	0.4564	0.7273	0.4564
No log	9.25	148	0.4582	0.7294	0.4582
No log	9.375	150	0.4572	0.7267	0.4572
No log	9.5	152	0.4559	0.7093	0.4559
No log	9.625	154	0.4566	0.7000	0.4566
No log	9.75	156	0.4582	0.6885	0.4582
No log	9.875	158	0.4588	0.6885	0.4588
No log	10.0	160	0.4588	0.6885	0.4588

Framework versions

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