ArabicNewSplits4_WithDuplicationsForScore5_FineTuningAraBERT_run2_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: 0.7003
• Qwk: 0.1902
• Mse: 0.7003
• Rmse: 0.8368

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	3.2724	0.0013	3.2724	1.8090
No log	0.5714	4	1.6404	-0.0370	1.6404	1.2808
No log	0.8571	6	0.7375	0.2487	0.7375	0.8588
No log	1.1429	8	0.7183	0.0863	0.7183	0.8475
No log	1.4286	10	1.0426	0.0462	1.0426	1.0211
No log	1.7143	12	0.6555	0.2644	0.6555	0.8096
No log	2.0	14	0.5599	-0.0159	0.5599	0.7483
No log	2.2857	16	0.6763	-0.0068	0.6763	0.8223
No log	2.5714	18	0.6929	-0.0133	0.6929	0.8324
No log	2.8571	20	0.6635	-0.0303	0.6635	0.8145
No log	3.1429	22	0.6053	-0.0081	0.6053	0.7780
No log	3.4286	24	0.5758	0.0476	0.5758	0.7588
No log	3.7143	26	0.5891	0.0388	0.5891	0.7675
No log	4.0	28	0.6103	0.0303	0.6103	0.7812
No log	4.2857	30	0.6981	-0.1733	0.6981	0.8355
No log	4.5714	32	0.6883	-0.1200	0.6883	0.8296
No log	4.8571	34	0.6776	-0.1200	0.6776	0.8231
No log	5.1429	36	0.6214	-0.0370	0.6214	0.7883
No log	5.4286	38	0.6182	0.0303	0.6182	0.7863
No log	5.7143	40	0.6389	-0.0233	0.6389	0.7993
No log	6.0	42	0.6574	-0.0233	0.6574	0.8108
No log	6.2857	44	0.6672	-0.0233	0.6672	0.8168
No log	6.5714	46	0.6705	-0.0303	0.6705	0.8188
No log	6.8571	48	0.6740	-0.0435	0.6740	0.8210
No log	7.1429	50	0.6679	-0.0435	0.6679	0.8173
No log	7.4286	52	0.6600	-0.0435	0.6600	0.8124
No log	7.7143	54	0.6606	0.0222	0.6606	0.8128
No log	8.0	56	0.6645	0.0222	0.6645	0.8152
No log	8.2857	58	0.6750	0.0897	0.6750	0.8216
No log	8.5714	60	0.6792	0.0897	0.6792	0.8241
No log	8.8571	62	0.6846	0.0692	0.6846	0.8274
No log	9.1429	64	0.6882	0.1304	0.6882	0.8296
No log	9.4286	66	0.6906	0.1392	0.6906	0.8310
No log	9.7143	68	0.6962	0.1392	0.6962	0.8344
No log	10.0	70	0.7003	0.1902	0.7003	0.8368

Framework versions

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