ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run1_AugV5_k1_task3_organization

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: 0.8599
• Qwk: 0.2536
• Mse: 0.8599
• Rmse: 0.9273

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: 8
• eval_batch_size: 8
• 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	Rmse
No log	0.2857	2	3.0823	0.0083	3.0823	1.7557
No log	0.5714	4	1.4234	0.0255	1.4234	1.1931
No log	0.8571	6	0.9207	0.0323	0.9207	0.9595
No log	1.1429	8	0.6004	0.0388	0.6004	0.7748
No log	1.4286	10	0.6162	0.0569	0.6162	0.7850
No log	1.7143	12	0.7470	0.2533	0.7470	0.8643
No log	2.0	14	0.6824	0.0720	0.6824	0.8261
No log	2.2857	16	0.5764	0.0	0.5764	0.7592
No log	2.5714	18	0.5637	0.0	0.5637	0.7508
No log	2.8571	20	0.5883	-0.0081	0.5883	0.7670
No log	3.1429	22	0.5974	-0.0081	0.5974	0.7729
No log	3.4286	24	0.5956	-0.0081	0.5956	0.7717
No log	3.7143	26	0.6185	0.0	0.6185	0.7865
No log	4.0	28	0.6257	0.0476	0.6257	0.7910
No log	4.2857	30	0.6127	0.0388	0.6127	0.7827
No log	4.5714	32	0.6522	-0.1701	0.6522	0.8076
No log	4.8571	34	0.6964	-0.0256	0.6964	0.8345
No log	5.1429	36	0.6876	-0.1200	0.6876	0.8292
No log	5.4286	38	0.7230	-0.0581	0.7230	0.8503
No log	5.7143	40	0.7597	0.0769	0.7597	0.8716
No log	6.0	42	0.7956	0.0939	0.7956	0.8920
No log	6.2857	44	0.8330	0.1304	0.8330	0.9127
No log	6.5714	46	0.8420	0.1398	0.8420	0.9176
No log	6.8571	48	0.8628	0.125	0.8628	0.9289
No log	7.1429	50	0.8697	0.125	0.8697	0.9326
No log	7.4286	52	0.8377	0.1064	0.8377	0.9153
No log	7.7143	54	0.8446	0.1220	0.8446	0.9190
No log	8.0	56	0.8455	0.1220	0.8455	0.9195
No log	8.2857	58	0.8407	0.1287	0.8407	0.9169
No log	8.5714	60	0.8263	0.1230	0.8263	0.9090
No log	8.8571	62	0.8376	0.0	0.8376	0.9152
No log	9.1429	64	0.8569	0.1005	0.8569	0.9257
No log	9.4286	66	0.8614	0.2150	0.8614	0.9281
No log	9.7143	68	0.8627	0.2536	0.8627	0.9288
No log	10.0	70	0.8599	0.2536	0.8599	0.9273

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1