ArabicNewSplits5_FineTuningAraBERT_run3_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: 0.8781
• Qwk: 0.0189
• Mse: 0.8781
• Rmse: 0.9371

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	0.2222	2	3.2006	-0.0138	3.2006	1.7890
No log	0.4444	4	1.6301	-0.0070	1.6301	1.2768
No log	0.6667	6	0.8751	0.1055	0.8751	0.9355
No log	0.8889	8	0.6365	0.1111	0.6365	0.7978
No log	1.1111	10	0.6867	0.0805	0.6867	0.8287
No log	1.3333	12	0.7228	0.0647	0.7228	0.8502
No log	1.5556	14	0.8281	0.0244	0.8281	0.9100
No log	1.7778	16	0.9924	0.0745	0.9924	0.9962
No log	2.0	18	0.9878	0.0038	0.9878	0.9939
No log	2.2222	20	0.7582	0.1644	0.7582	0.8707
No log	2.4444	22	0.5815	0.0569	0.5815	0.7625
No log	2.6667	24	0.5635	0.0569	0.5635	0.7507
No log	2.8889	26	0.5904	0.0222	0.5904	0.7684
No log	3.1111	28	0.7563	0.0667	0.7563	0.8697
No log	3.3333	30	0.7068	-0.0424	0.7068	0.8407
No log	3.5556	32	0.5768	0.0725	0.5768	0.7595
No log	3.7778	34	0.5994	0.1373	0.5994	0.7742
No log	4.0	36	0.6330	0.1345	0.6330	0.7956
No log	4.2222	38	0.5902	0.1795	0.5902	0.7682
No log	4.4444	40	0.7300	-0.0058	0.7300	0.8544
No log	4.6667	42	0.8055	-0.0169	0.8055	0.8975
No log	4.8889	44	0.7314	-0.0058	0.7314	0.8552
No log	5.1111	46	0.6720	-0.0133	0.6720	0.8197
No log	5.3333	48	0.6861	0.0604	0.6861	0.8283
No log	5.5556	50	0.6872	0.0728	0.6872	0.8290
No log	5.7778	52	0.7147	0.0	0.7147	0.8454
No log	6.0	54	0.7431	0.0	0.7431	0.8620
No log	6.2222	56	0.7908	0.0409	0.7908	0.8893
No log	6.4444	58	0.8037	0.0118	0.8037	0.8965
No log	6.6667	60	0.8214	0.0909	0.8214	0.9063
No log	6.8889	62	0.8139	0.0909	0.8139	0.9021
No log	7.1111	64	0.7929	0.0703	0.7929	0.8905
No log	7.3333	66	0.7879	-0.0105	0.7879	0.8876
No log	7.5556	68	0.7815	-0.0105	0.7815	0.8840
No log	7.7778	70	0.7996	0.0457	0.7996	0.8942
No log	8.0	72	0.8233	0.0103	0.8233	0.9074
No log	8.2222	74	0.8208	0.0103	0.8208	0.9060
No log	8.4444	76	0.8233	0.0457	0.8233	0.9074
No log	8.6667	78	0.8354	-0.0297	0.8354	0.9140
No log	8.8889	80	0.8551	0.0495	0.8551	0.9247
No log	9.1111	82	0.8683	0.0049	0.8683	0.9318
No log	9.3333	84	0.8773	0.0049	0.8773	0.9366
No log	9.5556	86	0.8796	0.0531	0.8796	0.9379
No log	9.7778	88	0.8787	0.0189	0.8787	0.9374
No log	10.0	90	0.8781	0.0189	0.8781	0.9371

Framework versions

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