ArabicNewSplits2_FineTuningAraBERT_noAug_task1_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: 1.1149
• Qwk: 0.4220
• Mse: 1.1149
• Rmse: 1.0559

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	0.5	2	5.1271	0.0848	5.1271	2.2643
No log	1.0	4	3.3027	0.0056	3.3027	1.8173
No log	1.5	6	2.2685	0.1324	2.2685	1.5062
No log	2.0	8	1.5495	0.0669	1.5495	1.2448
No log	2.5	10	1.3312	0.1321	1.3312	1.1538
No log	3.0	12	1.1926	0.3233	1.1926	1.0921
No log	3.5	14	1.1219	0.2410	1.1219	1.0592
No log	4.0	16	1.1902	0.1619	1.1902	1.0910
No log	4.5	18	1.3087	0.0816	1.3087	1.1440
No log	5.0	20	1.3486	0.0498	1.3486	1.1613
No log	5.5	22	1.3088	0.3211	1.3088	1.1440
No log	6.0	24	1.2636	0.4338	1.2636	1.1241
No log	6.5	26	1.2286	0.3687	1.2286	1.1084
No log	7.0	28	1.1642	0.3687	1.1642	1.0790
No log	7.5	30	1.1134	0.4031	1.1134	1.0552
No log	8.0	32	1.1224	0.4505	1.1224	1.0594
No log	8.5	34	1.1062	0.4791	1.1062	1.0518
No log	9.0	36	1.1067	0.4498	1.1067	1.0520
No log	9.5	38	1.1121	0.4498	1.1121	1.0546
No log	10.0	40	1.1149	0.4220	1.1149	1.0559

Framework versions

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