arabert_cross_organization_task4_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: 0.5515
• Qwk: 0.8285
• Mse: 0.5515

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	0.1176	2	1.8640	0.0713	1.8640
No log	0.2353	4	1.2390	0.2961	1.2390
No log	0.3529	6	1.0659	0.5023	1.0659
No log	0.4706	8	0.7195	0.6388	0.7195
No log	0.5882	10	0.6736	0.6636	0.6736
No log	0.7059	12	0.6560	0.7539	0.6560
No log	0.8235	14	0.6088	0.7149	0.6088
No log	0.9412	16	0.5695	0.7426	0.5695
No log	1.0588	18	0.5646	0.7706	0.5646
No log	1.1765	20	0.5616	0.7796	0.5616
No log	1.2941	22	0.5516	0.7811	0.5516
No log	1.4118	24	0.5475	0.7868	0.5475
No log	1.5294	26	0.5246	0.7436	0.5246
No log	1.6471	28	0.5463	0.6820	0.5463
No log	1.7647	30	0.5511	0.7714	0.5511
No log	1.8824	32	0.6634	0.7890	0.6634
No log	2.0	34	0.5815	0.7811	0.5815
No log	2.1176	36	0.5186	0.7320	0.5186
No log	2.2353	38	0.5161	0.7348	0.5161
No log	2.3529	40	0.5145	0.7637	0.5145
No log	2.4706	42	0.5367	0.7805	0.5367
No log	2.5882	44	0.5238	0.7807	0.5238
No log	2.7059	46	0.5169	0.7771	0.5169
No log	2.8235	48	0.5287	0.7852	0.5287
No log	2.9412	50	0.5446	0.7867	0.5446
No log	3.0588	52	0.5892	0.7786	0.5892
No log	3.1765	54	0.5913	0.7818	0.5913
No log	3.2941	56	0.5352	0.7813	0.5352
No log	3.4118	58	0.5197	0.7785	0.5197
No log	3.5294	60	0.5895	0.7951	0.5895
No log	3.6471	62	0.6125	0.7950	0.6125
No log	3.7647	64	0.5768	0.7875	0.5768
No log	3.8824	66	0.5110	0.7717	0.5110
No log	4.0	68	0.5375	0.7910	0.5375
No log	4.1176	70	0.6011	0.7996	0.6011
No log	4.2353	72	0.5525	0.7824	0.5525
No log	4.3529	74	0.5620	0.7907	0.5620
No log	4.4706	76	0.5220	0.7908	0.5220
No log	4.5882	78	0.5028	0.7641	0.5028
No log	4.7059	80	0.5102	0.7841	0.5102
No log	4.8235	82	0.6101	0.8166	0.6101
No log	4.9412	84	0.7637	0.8377	0.7637
No log	5.0588	86	0.6895	0.8271	0.6895
No log	5.1765	88	0.5154	0.7893	0.5154
No log	5.2941	90	0.4808	0.7932	0.4808
No log	5.4118	92	0.5241	0.7847	0.5241
No log	5.5294	94	0.6303	0.8282	0.6303
No log	5.6471	96	0.6026	0.8213	0.6026
No log	5.7647	98	0.5078	0.8217	0.5078
No log	5.8824	100	0.4940	0.8070	0.4940
No log	6.0	102	0.5633	0.8182	0.5633
No log	6.1176	104	0.7172	0.8375	0.7172
No log	6.2353	106	0.7195	0.8305	0.7195
No log	6.3529	108	0.5965	0.8196	0.5965
No log	6.4706	110	0.5114	0.7928	0.5114
No log	6.5882	112	0.5064	0.7886	0.5064
No log	6.7059	114	0.5439	0.8123	0.5439
No log	6.8235	116	0.6118	0.8257	0.6118
No log	6.9412	118	0.6143	0.8179	0.6143
No log	7.0588	120	0.6140	0.8174	0.6140
No log	7.1765	122	0.5766	0.8207	0.5766
No log	7.2941	124	0.5369	0.8188	0.5369
No log	7.4118	126	0.5480	0.8132	0.5480
No log	7.5294	128	0.6078	0.8236	0.6078
No log	7.6471	130	0.6701	0.8235	0.6701
No log	7.7647	132	0.6414	0.8221	0.6414
No log	7.8824	134	0.5864	0.8195	0.5864
No log	8.0	136	0.5473	0.8186	0.5473
No log	8.1176	138	0.5345	0.8188	0.5345
No log	8.2353	140	0.5432	0.8202	0.5432
No log	8.3529	142	0.5420	0.8239	0.5420
No log	8.4706	144	0.5381	0.8188	0.5381
No log	8.5882	146	0.5139	0.8016	0.5139
No log	8.7059	148	0.5052	0.7981	0.5052
No log	8.8235	150	0.5095	0.7948	0.5095
No log	8.9412	152	0.5334	0.8173	0.5334
No log	9.0588	154	0.5783	0.8340	0.5783
No log	9.1765	156	0.6036	0.8366	0.6036
No log	9.2941	158	0.6056	0.8403	0.6056
No log	9.4118	160	0.5911	0.8320	0.5911
No log	9.5294	162	0.5683	0.8301	0.5683
No log	9.6471	164	0.5583	0.8287	0.5583
No log	9.7647	166	0.5530	0.8269	0.5530
No log	9.8824	168	0.5512	0.8285	0.5512
No log	10.0	170	0.5515	0.8285	0.5515

Framework versions

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