ArabicNewSplits6_FineTuningAraBERT_run1_AugV5_k3_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: 0.6459
• Qwk: 0.3462
• Mse: 0.6459
• Rmse: 0.8037

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	3.2209	-0.0149	3.2209	1.7947
No log	0.25	4	1.6372	-0.0370	1.6372	1.2795
No log	0.375	6	0.9690	0.0114	0.9690	0.9844
No log	0.5	8	1.0805	0.1525	1.0805	1.0395
No log	0.625	10	0.8810	0.1652	0.8810	0.9386
No log	0.75	12	0.5561	0.0769	0.5561	0.7457
No log	0.875	14	0.5243	0.0769	0.5243	0.7241
No log	1.0	16	0.9632	0.0894	0.9632	0.9814
No log	1.125	18	1.1586	0.0698	1.1586	1.0764
No log	1.25	20	0.6662	0.1746	0.6662	0.8162
No log	1.375	22	0.5887	-0.0068	0.5887	0.7673
No log	1.5	24	0.5635	0.0145	0.5635	0.7507
No log	1.625	26	0.5323	0.1304	0.5323	0.7296
No log	1.75	28	0.6102	0.3333	0.6102	0.7812
No log	1.875	30	0.5598	0.2994	0.5598	0.7482
No log	2.0	32	0.4928	0.1407	0.4928	0.7020
No log	2.125	34	0.5447	0.2000	0.5447	0.7380
No log	2.25	36	0.4847	0.1592	0.4847	0.6962
No log	2.375	38	0.8063	0.2676	0.8063	0.8979
No log	2.5	40	0.9362	0.1795	0.9362	0.9676
No log	2.625	42	0.6269	0.2000	0.6269	0.7918
No log	2.75	44	0.5263	0.2795	0.5263	0.7255
No log	2.875	46	0.6123	0.1899	0.6123	0.7825
No log	3.0	48	0.6267	0.2393	0.6267	0.7916
No log	3.125	50	0.5635	0.2096	0.5635	0.7506
No log	3.25	52	0.5925	0.1902	0.5925	0.7698
No log	3.375	54	0.5840	0.1788	0.5840	0.7642
No log	3.5	56	0.6058	0.1111	0.6058	0.7784
No log	3.625	58	0.6488	0.2381	0.6488	0.8055
No log	3.75	60	0.6515	0.2832	0.6515	0.8071
No log	3.875	62	0.5797	0.1801	0.5797	0.7614
No log	4.0	64	0.6910	0.1568	0.6910	0.8312
No log	4.125	66	0.6295	0.2644	0.6295	0.7934
No log	4.25	68	0.7177	0.2871	0.7177	0.8472
No log	4.375	70	0.8754	0.1545	0.8754	0.9356
No log	4.5	72	0.6764	0.3171	0.6764	0.8225
No log	4.625	74	0.6746	0.2692	0.6746	0.8213
No log	4.75	76	0.7712	0.2857	0.7712	0.8782
No log	4.875	78	0.5993	0.3299	0.5993	0.7741
No log	5.0	80	0.7730	0.1538	0.7730	0.8792
No log	5.125	82	0.9466	0.1686	0.9466	0.9730
No log	5.25	84	0.7720	0.1443	0.7720	0.8786
No log	5.375	86	0.6177	0.3333	0.6177	0.7859
No log	5.5	88	0.6266	0.3663	0.6266	0.7916
No log	5.625	90	0.6442	0.28	0.6442	0.8026
No log	5.75	92	0.6346	0.2893	0.6346	0.7966
No log	5.875	94	0.5630	0.3508	0.5630	0.7503
No log	6.0	96	0.5251	0.3118	0.5251	0.7246
No log	6.125	98	0.5441	0.3878	0.5441	0.7376
No log	6.25	100	0.5865	0.3535	0.5865	0.7658
No log	6.375	102	0.6448	0.4133	0.6448	0.8030
No log	6.5	104	0.5700	0.3301	0.5700	0.7550
No log	6.625	106	0.5431	0.3237	0.5431	0.7369
No log	6.75	108	0.5494	0.3237	0.5494	0.7412
No log	6.875	110	0.6343	0.4035	0.6343	0.7965
No log	7.0	112	0.7333	0.3833	0.7333	0.8563
No log	7.125	114	0.7240	0.3739	0.7240	0.8509
No log	7.25	116	0.6534	0.3333	0.6534	0.8083
No log	7.375	118	0.7050	0.3052	0.7050	0.8396
No log	7.5	120	0.7862	0.3739	0.7862	0.8867
No log	7.625	122	0.8553	0.2203	0.8553	0.9248
No log	7.75	124	0.8945	0.1673	0.8945	0.9458
No log	7.875	126	0.7728	0.3128	0.7728	0.8791
No log	8.0	128	0.6123	0.4118	0.6123	0.7825
No log	8.125	130	0.5752	0.3951	0.5752	0.7584
No log	8.25	132	0.5779	0.4118	0.5779	0.7602
No log	8.375	134	0.6106	0.3786	0.6106	0.7814
No log	8.5	136	0.6981	0.3143	0.6981	0.8355
No log	8.625	138	0.8229	0.2821	0.8229	0.9072
No log	8.75	140	0.8856	0.2846	0.8856	0.9410
No log	8.875	142	0.8606	0.2900	0.8606	0.9277
No log	9.0	144	0.7753	0.3188	0.7753	0.8805
No log	9.125	146	0.6760	0.3427	0.6760	0.8222
No log	9.25	148	0.6357	0.3462	0.6357	0.7973
No log	9.375	150	0.6122	0.3365	0.6122	0.7825
No log	9.5	152	0.6067	0.3365	0.6067	0.7789
No log	9.625	154	0.6176	0.3365	0.6176	0.7859
No log	9.75	156	0.6340	0.3365	0.6340	0.7962
No log	9.875	158	0.6411	0.3462	0.6411	0.8007
No log	10.0	160	0.6459	0.3462	0.6459	0.8037

Framework versions

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