ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run3_AugV5_k3_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: 0.6317
• Qwk: 0.7478
• Mse: 0.6317
• Rmse: 0.7948

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	0.1053	2	5.0300	0.0104	5.0300	2.2428
No log	0.2105	4	3.4484	0.0816	3.4484	1.8570
No log	0.3158	6	1.9537	0.1429	1.9537	1.3978
No log	0.4211	8	1.3386	0.2182	1.3386	1.1570
No log	0.5263	10	1.0294	0.2894	1.0294	1.0146
No log	0.6316	12	1.0124	0.2423	1.0124	1.0062
No log	0.7368	14	0.9383	0.3075	0.9383	0.9687
No log	0.8421	16	0.8835	0.3797	0.8835	0.9400
No log	0.9474	18	0.8324	0.5009	0.8324	0.9123
No log	1.0526	20	0.7614	0.5091	0.7614	0.8726
No log	1.1579	22	1.0157	0.5010	1.0157	1.0078
No log	1.2632	24	1.3209	0.4284	1.3209	1.1493
No log	1.3684	26	1.4267	0.3942	1.4267	1.1944
No log	1.4737	28	0.9740	0.6013	0.9740	0.9869
No log	1.5789	30	0.7838	0.7408	0.7838	0.8853
No log	1.6842	32	0.7764	0.7515	0.7764	0.8811
No log	1.7895	34	1.2608	0.4906	1.2608	1.1228
No log	1.8947	36	1.7894	0.3812	1.7894	1.3377
No log	2.0	38	1.3946	0.4635	1.3946	1.1809
No log	2.1053	40	1.0695	0.5954	1.0695	1.0342
No log	2.2105	42	0.6496	0.7285	0.6496	0.8060
No log	2.3158	44	0.5928	0.7639	0.5928	0.7699
No log	2.4211	46	0.6044	0.7396	0.6044	0.7774
No log	2.5263	48	0.8918	0.6386	0.8918	0.9443
No log	2.6316	50	1.0785	0.5714	1.0785	1.0385
No log	2.7368	52	0.8498	0.6660	0.8498	0.9219
No log	2.8421	54	0.6111	0.7610	0.6111	0.7818
No log	2.9474	56	0.6701	0.7572	0.6701	0.8186
No log	3.0526	58	0.8330	0.7221	0.8330	0.9127
No log	3.1579	60	0.7739	0.7365	0.7739	0.8797
No log	3.2632	62	0.5611	0.7645	0.5611	0.7491
No log	3.3684	64	0.5699	0.7400	0.5699	0.7549
No log	3.4737	66	0.7233	0.6551	0.7233	0.8505
No log	3.5789	68	0.7040	0.6746	0.7040	0.8391
No log	3.6842	70	0.6718	0.7047	0.6718	0.8196
No log	3.7895	72	0.5946	0.7654	0.5946	0.7711
No log	3.8947	74	0.6128	0.7723	0.6128	0.7828
No log	4.0	76	0.6559	0.7495	0.6559	0.8098
No log	4.1053	78	0.6436	0.7633	0.6436	0.8023
No log	4.2105	80	0.6010	0.7669	0.6010	0.7752
No log	4.3158	82	0.6147	0.7698	0.6147	0.7840
No log	4.4211	84	0.6468	0.7278	0.6468	0.8043
No log	4.5263	86	0.6148	0.7620	0.6148	0.7841
No log	4.6316	88	0.6005	0.7632	0.6005	0.7749
No log	4.7368	90	0.6770	0.7563	0.6770	0.8228
No log	4.8421	92	0.7320	0.7452	0.7320	0.8556
No log	4.9474	94	0.6612	0.7549	0.6612	0.8131
No log	5.0526	96	0.5885	0.7754	0.5885	0.7672
No log	5.1579	98	0.6548	0.7387	0.6548	0.8092
No log	5.2632	100	0.7356	0.7196	0.7356	0.8577
No log	5.3684	102	0.6760	0.7487	0.6760	0.8222
No log	5.4737	104	0.6136	0.7668	0.6136	0.7833
No log	5.5789	106	0.6918	0.7498	0.6918	0.8318
No log	5.6842	108	0.7124	0.7606	0.7124	0.8441
No log	5.7895	110	0.6328	0.7568	0.6328	0.7955
No log	5.8947	112	0.5844	0.7743	0.5844	0.7645
No log	6.0	114	0.5792	0.7563	0.5792	0.7611
No log	6.1053	116	0.5730	0.7562	0.5730	0.7570
No log	6.2105	118	0.5786	0.7831	0.5786	0.7607
No log	6.3158	120	0.5908	0.7732	0.5908	0.7686
No log	6.4211	122	0.6077	0.7802	0.6077	0.7795
No log	6.5263	124	0.6195	0.7592	0.6195	0.7871
No log	6.6316	126	0.6185	0.7528	0.6185	0.7864
No log	6.7368	128	0.6298	0.7573	0.6298	0.7936
No log	6.8421	130	0.6458	0.7571	0.6458	0.8036
No log	6.9474	132	0.6495	0.7646	0.6495	0.8059
No log	7.0526	134	0.6192	0.7560	0.6192	0.7869
No log	7.1579	136	0.5980	0.7692	0.5980	0.7733
No log	7.2632	138	0.6057	0.7682	0.6057	0.7783
No log	7.3684	140	0.6084	0.7601	0.6084	0.7800
No log	7.4737	142	0.6264	0.7730	0.6264	0.7914
No log	7.5789	144	0.6320	0.7734	0.6320	0.7950
No log	7.6842	146	0.6182	0.7617	0.6182	0.7862
No log	7.7895	148	0.6150	0.7833	0.6150	0.7842
No log	7.8947	150	0.6255	0.7766	0.6255	0.7909
No log	8.0	152	0.6424	0.7657	0.6424	0.8015
No log	8.1053	154	0.6638	0.7666	0.6638	0.8147
No log	8.2105	156	0.6642	0.7677	0.6642	0.8150
No log	8.3158	158	0.6563	0.7760	0.6563	0.8101
No log	8.4211	160	0.6566	0.7651	0.6566	0.8103
No log	8.5263	162	0.6603	0.7460	0.6603	0.8126
No log	8.6316	164	0.6609	0.7419	0.6609	0.8130
No log	8.7368	166	0.6578	0.7605	0.6578	0.8110
No log	8.8421	168	0.6457	0.7608	0.6457	0.8035
No log	8.9474	170	0.6329	0.7563	0.6329	0.7956
No log	9.0526	172	0.6259	0.7513	0.6259	0.7912
No log	9.1579	174	0.6230	0.7480	0.6230	0.7893
No log	9.2632	176	0.6248	0.7480	0.6248	0.7904
No log	9.3684	178	0.6322	0.7563	0.6322	0.7951
No log	9.4737	180	0.6375	0.7560	0.6375	0.7985
No log	9.5789	182	0.6394	0.7608	0.6394	0.7996
No log	9.6842	184	0.6379	0.7476	0.6379	0.7987
No log	9.7895	186	0.6355	0.7478	0.6355	0.7972
No log	9.8947	188	0.6329	0.7478	0.6329	0.7955
No log	10.0	190	0.6317	0.7478	0.6317	0.7948

Framework versions

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