ArabicNewSplits5_FineTuningAraBERT_run2_AugV5_k1_task5_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: 0.7991
• Qwk: 0.6298
• Mse: 0.7991
• Rmse: 0.8939

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	0.25	2	2.2724	0.0239	2.2724	1.5074
No log	0.5	4	1.5637	0.1917	1.5637	1.2505
No log	0.75	6	1.2804	0.2634	1.2804	1.1316
No log	1.0	8	1.4096	0.2049	1.4096	1.1872
No log	1.25	10	1.6595	0.3169	1.6595	1.2882
No log	1.5	12	1.7883	0.2853	1.7883	1.3373
No log	1.75	14	1.6356	0.2749	1.6356	1.2789
No log	2.0	16	1.4498	0.0928	1.4498	1.2041
No log	2.25	18	1.4479	0.1548	1.4479	1.2033
No log	2.5	20	1.3916	0.2186	1.3916	1.1797
No log	2.75	22	1.3096	0.1682	1.3096	1.1444
No log	3.0	24	1.2948	0.1071	1.2948	1.1379
No log	3.25	26	1.3057	0.1071	1.3057	1.1427
No log	3.5	28	1.3117	0.1232	1.3117	1.1453
No log	3.75	30	1.3316	0.2487	1.3316	1.1539
No log	4.0	32	1.3250	0.3480	1.3250	1.1511
No log	4.25	34	1.2770	0.3261	1.2770	1.1301
No log	4.5	36	1.2126	0.2111	1.2126	1.1012
No log	4.75	38	1.1552	0.2268	1.1552	1.0748
No log	5.0	40	1.1145	0.2689	1.1145	1.0557
No log	5.25	42	1.0775	0.3010	1.0775	1.0380
No log	5.5	44	1.0556	0.3548	1.0556	1.0274
No log	5.75	46	1.0201	0.4132	1.0201	1.0100
No log	6.0	48	0.9952	0.4514	0.9952	0.9976
No log	6.25	50	0.9735	0.4814	0.9735	0.9867
No log	6.5	52	0.9374	0.5062	0.9374	0.9682
No log	6.75	54	0.9078	0.5131	0.9078	0.9528
No log	7.0	56	0.8908	0.5254	0.8908	0.9438
No log	7.25	58	0.8742	0.5490	0.8742	0.9350
No log	7.5	60	0.8482	0.5655	0.8482	0.9210
No log	7.75	62	0.8423	0.5772	0.8423	0.9178
No log	8.0	64	0.8649	0.5652	0.8649	0.9300
No log	8.25	66	0.8831	0.5812	0.8831	0.9397
No log	8.5	68	0.8769	0.5812	0.8769	0.9364
No log	8.75	70	0.8455	0.6106	0.8455	0.9195
No log	9.0	72	0.8144	0.6338	0.8144	0.9024
No log	9.25	74	0.8078	0.6338	0.8078	0.8988
No log	9.5	76	0.8050	0.6263	0.8050	0.8972
No log	9.75	78	0.8004	0.6280	0.8004	0.8947
No log	10.0	80	0.7991	0.6298	0.7991	0.8939

Framework versions

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