ArabicNewSplits5_FineTuningAraBERT_run2_AugV5_k2_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: 0.7955
• Qwk: 0.6859
• Mse: 0.7955
• Rmse: 0.8919

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	0.125	2	5.1059	-0.0238	5.1059	2.2596
No log	0.25	4	2.9393	0.0863	2.9393	1.7144
No log	0.375	6	1.8752	0.1210	1.8752	1.3694
No log	0.5	8	1.6197	0.1017	1.6197	1.2727
No log	0.625	10	1.9296	-0.0344	1.9296	1.3891
No log	0.75	12	1.8116	-0.1098	1.8116	1.3459
No log	0.875	14	2.1000	-0.1223	2.1000	1.4491
No log	1.0	16	2.1515	-0.0859	2.1515	1.4668
No log	1.125	18	2.5569	-0.0317	2.5569	1.5990
No log	1.25	20	1.8494	0.0290	1.8494	1.3599
No log	1.375	22	1.1987	0.2562	1.1987	1.0948
No log	1.5	24	1.1159	0.3640	1.1159	1.0563
No log	1.625	26	1.1712	0.3681	1.1712	1.0822
No log	1.75	28	1.1964	0.3576	1.1964	1.0938
No log	1.875	30	1.1917	0.3201	1.1917	1.0917
No log	2.0	32	1.3469	0.1222	1.3469	1.1605
No log	2.125	34	1.6531	0.0540	1.6531	1.2857
No log	2.25	36	1.6781	0.1355	1.6781	1.2954
No log	2.375	38	1.3813	0.1723	1.3813	1.1753
No log	2.5	40	1.0940	0.3416	1.0940	1.0460
No log	2.625	42	1.0161	0.3882	1.0161	1.0080
No log	2.75	44	1.0008	0.4038	1.0008	1.0004
No log	2.875	46	1.0746	0.3615	1.0746	1.0367
No log	3.0	48	1.0037	0.4337	1.0037	1.0019
No log	3.125	50	0.8813	0.4934	0.8813	0.9388
No log	3.25	52	0.8219	0.4930	0.8219	0.9066
No log	3.375	54	0.8188	0.4842	0.8188	0.9049
No log	3.5	56	0.7809	0.5433	0.7809	0.8837
No log	3.625	58	0.7510	0.5510	0.7510	0.8666
No log	3.75	60	0.9111	0.5707	0.9111	0.9545
No log	3.875	62	1.4708	0.4556	1.4708	1.2128
No log	4.0	64	1.6728	0.4538	1.6728	1.2934
No log	4.125	66	1.4221	0.4596	1.4221	1.1925
No log	4.25	68	1.0266	0.6027	1.0266	1.0132
No log	4.375	70	0.7413	0.6308	0.7413	0.8610
No log	4.5	72	0.6523	0.6924	0.6523	0.8076
No log	4.625	74	0.6940	0.6997	0.6940	0.8330
No log	4.75	76	0.7507	0.6364	0.7507	0.8664
No log	4.875	78	0.7390	0.6021	0.7390	0.8597
No log	5.0	80	0.7044	0.6501	0.7044	0.8393
No log	5.125	82	0.6646	0.6428	0.6646	0.8152
No log	5.25	84	0.6654	0.6671	0.6654	0.8157
No log	5.375	86	0.6971	0.6339	0.6971	0.8349
No log	5.5	88	0.6896	0.6264	0.6896	0.8304
No log	5.625	90	0.6604	0.6501	0.6604	0.8126
No log	5.75	92	0.6874	0.6969	0.6874	0.8291
No log	5.875	94	0.7291	0.7012	0.7291	0.8539
No log	6.0	96	0.7497	0.6711	0.7497	0.8658
No log	6.125	98	0.7371	0.6993	0.7371	0.8586
No log	6.25	100	0.7079	0.6742	0.7079	0.8414
No log	6.375	102	0.7039	0.7041	0.7039	0.8390
No log	6.5	104	0.7211	0.6852	0.7211	0.8492
No log	6.625	106	0.7157	0.6945	0.7157	0.8460
No log	6.75	108	0.7312	0.7117	0.7312	0.8551
No log	6.875	110	0.7477	0.7181	0.7477	0.8647
No log	7.0	112	0.7433	0.7181	0.7433	0.8622
No log	7.125	114	0.7286	0.7151	0.7286	0.8536
No log	7.25	116	0.7257	0.7187	0.7257	0.8519
No log	7.375	118	0.7132	0.7158	0.7132	0.8445
No log	7.5	120	0.7199	0.7210	0.7199	0.8484
No log	7.625	122	0.7381	0.6968	0.7381	0.8591
No log	7.75	124	0.7452	0.6956	0.7452	0.8632
No log	7.875	126	0.7650	0.6751	0.7650	0.8747
No log	8.0	128	0.7856	0.6654	0.7856	0.8863
No log	8.125	130	0.7923	0.6647	0.7923	0.8901
No log	8.25	132	0.7799	0.6629	0.7799	0.8831
No log	8.375	134	0.7776	0.6693	0.7776	0.8818
No log	8.5	136	0.7540	0.6820	0.7540	0.8683
No log	8.625	138	0.7308	0.6838	0.7308	0.8549
No log	8.75	140	0.7165	0.6859	0.7165	0.8465
No log	8.875	142	0.7172	0.6859	0.7172	0.8469
No log	9.0	144	0.7227	0.6919	0.7227	0.8501
No log	9.125	146	0.7270	0.6919	0.7270	0.8526
No log	9.25	148	0.7373	0.7050	0.7373	0.8587
No log	9.375	150	0.7509	0.7031	0.7509	0.8665
No log	9.5	152	0.7659	0.7012	0.7659	0.8751
No log	9.625	154	0.7805	0.6859	0.7805	0.8835
No log	9.75	156	0.7887	0.6859	0.7887	0.8881
No log	9.875	158	0.7930	0.6859	0.7930	0.8905
No log	10.0	160	0.7955	0.6859	0.7955	0.8919

Framework versions

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