ArabicNewSplits5_FineTuningAraBERT_run2_AugV5_k2_task2_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: 0.8614
• Qwk: 0.5361
• Mse: 0.8614
• Rmse: 0.9281

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	3.8515	0.0002	3.8515	1.9625
No log	0.25	4	2.3096	0.1010	2.3096	1.5197
No log	0.375	6	1.3981	0.1663	1.3981	1.1824
No log	0.5	8	0.9042	0.0135	0.9042	0.9509
No log	0.625	10	0.7606	0.1087	0.7606	0.8721
No log	0.75	12	0.6881	0.1505	0.6881	0.8295
No log	0.875	14	0.6893	0.2168	0.6893	0.8302
No log	1.0	16	0.7271	0.1526	0.7271	0.8527
No log	1.125	18	0.8376	0.0417	0.8376	0.9152
No log	1.25	20	0.8475	0.1569	0.8475	0.9206
No log	1.375	22	1.1247	0.0739	1.1247	1.0605
No log	1.5	24	1.3601	0.1119	1.3601	1.1662
No log	1.625	26	1.2206	0.1344	1.2206	1.1048
No log	1.75	28	1.4393	0.1194	1.4393	1.1997
No log	1.875	30	1.6168	0.1392	1.6168	1.2715
No log	2.0	32	1.3110	0.1359	1.3110	1.1450
No log	2.125	34	0.8776	0.1501	0.8776	0.9368
No log	2.25	36	0.6537	0.3391	0.6537	0.8085
No log	2.375	38	0.5958	0.3663	0.5958	0.7719
No log	2.5	40	0.6111	0.3731	0.6111	0.7817
No log	2.625	42	0.7253	0.2800	0.7253	0.8516
No log	2.75	44	0.9967	0.0866	0.9967	0.9984
No log	2.875	46	1.0849	0.1120	1.0849	1.0416
No log	3.0	48	0.9059	0.2173	0.9059	0.9518
No log	3.125	50	0.7349	0.3365	0.7349	0.8572
No log	3.25	52	0.6563	0.3622	0.6563	0.8101
No log	3.375	54	0.5934	0.4205	0.5934	0.7703
No log	3.5	56	0.5982	0.3953	0.5982	0.7735
No log	3.625	58	0.6958	0.3981	0.6958	0.8342
No log	3.75	60	0.9557	0.3765	0.9557	0.9776
No log	3.875	62	1.1315	0.2616	1.1315	1.0637
No log	4.0	64	1.1604	0.2842	1.1604	1.0772
No log	4.125	66	1.0819	0.2985	1.0819	1.0401
No log	4.25	68	0.9620	0.3607	0.9620	0.9808
No log	4.375	70	0.7088	0.4919	0.7088	0.8419
No log	4.5	72	0.5833	0.4512	0.5833	0.7638
No log	4.625	74	0.5693	0.4532	0.5693	0.7545
No log	4.75	76	0.5996	0.4974	0.5996	0.7743
No log	4.875	78	0.6358	0.5211	0.6358	0.7974
No log	5.0	80	0.7175	0.5153	0.7175	0.8471
No log	5.125	82	0.6970	0.5544	0.6970	0.8349
No log	5.25	84	0.7384	0.5205	0.7384	0.8593
No log	5.375	86	0.7783	0.4982	0.7783	0.8822
No log	5.5	88	0.7563	0.5223	0.7563	0.8696
No log	5.625	90	0.6762	0.5840	0.6762	0.8223
No log	5.75	92	0.6589	0.5353	0.6589	0.8117
No log	5.875	94	0.6628	0.5225	0.6628	0.8141
No log	6.0	96	0.6810	0.5616	0.6810	0.8252
No log	6.125	98	0.7192	0.5865	0.7192	0.8480
No log	6.25	100	0.7691	0.5545	0.7691	0.8770
No log	6.375	102	0.7938	0.5420	0.7938	0.8909
No log	6.5	104	0.7961	0.5572	0.7961	0.8922
No log	6.625	106	0.8108	0.5522	0.8108	0.9004
No log	6.75	108	0.8078	0.5661	0.8078	0.8988
No log	6.875	110	0.7875	0.5778	0.7875	0.8874
No log	7.0	112	0.7790	0.5432	0.7790	0.8826
No log	7.125	114	0.7856	0.5285	0.7856	0.8863
No log	7.25	116	0.8066	0.5308	0.8066	0.8981
No log	7.375	118	0.8118	0.5381	0.8118	0.9010
No log	7.5	120	0.8107	0.5461	0.8107	0.9004
No log	7.625	122	0.8194	0.5581	0.8194	0.9052
No log	7.75	124	0.8492	0.5583	0.8492	0.9215
No log	7.875	126	0.8828	0.5303	0.8828	0.9396
No log	8.0	128	0.8739	0.5717	0.8739	0.9348
No log	8.125	130	0.8609	0.5460	0.8609	0.9279
No log	8.25	132	0.8632	0.5689	0.8632	0.9291
No log	8.375	134	0.8695	0.5581	0.8695	0.9325
No log	8.5	136	0.8763	0.5713	0.8763	0.9361
No log	8.625	138	0.8746	0.5749	0.8746	0.9352
No log	8.75	140	0.8675	0.5543	0.8675	0.9314
No log	8.875	142	0.8687	0.5466	0.8687	0.9320
No log	9.0	144	0.8789	0.5443	0.8789	0.9375
No log	9.125	146	0.8830	0.5489	0.8830	0.9397
No log	9.25	148	0.8830	0.5490	0.8830	0.9397
No log	9.375	150	0.8795	0.5514	0.8795	0.9378
No log	9.5	152	0.8733	0.5626	0.8733	0.9345
No log	9.625	154	0.8667	0.5429	0.8667	0.9309
No log	9.75	156	0.8628	0.5431	0.8628	0.9289
No log	9.875	158	0.8615	0.5457	0.8615	0.9282
No log	10.0	160	0.8614	0.5361	0.8614	0.9281

Framework versions

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