ArabicNewSplits_FineTuningAraBERT_AugV5_k4_task2_organization_fold1

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: 1.0147
• Qwk: 0.1665
• Mse: 1.0147
• Rmse: 1.0073

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	3.8832	0.0941	3.8832	1.9706
No log	0.25	4	2.5260	0.1416	2.5260	1.5893
No log	0.375	6	1.0806	0.3274	1.0806	1.0395
No log	0.5	8	0.8313	0.0534	0.8313	0.9118
No log	0.625	10	0.9007	-0.0182	0.9007	0.9490
No log	0.75	12	0.7833	0.0365	0.7833	0.8850
No log	0.875	14	0.8533	-0.0556	0.8533	0.9237
No log	1.0	16	0.9062	0.1038	0.9062	0.9520
No log	1.125	18	0.8667	0.1038	0.8667	0.9310
No log	1.25	20	0.8635	0.1038	0.8635	0.9293
No log	1.375	22	0.9500	0.1038	0.9500	0.9747
No log	1.5	24	1.0484	0.0120	1.0484	1.0239
No log	1.625	26	1.0764	-0.0405	1.0764	1.0375
No log	1.75	28	1.0692	-0.0405	1.0692	1.0340
No log	1.875	30	0.9087	-0.0205	0.9087	0.9533
No log	2.0	32	0.8329	-0.0556	0.8329	0.9126
No log	2.125	34	0.8125	-0.0556	0.8125	0.9014
No log	2.25	36	0.8017	-0.0556	0.8017	0.8954
No log	2.375	38	0.8288	-0.0556	0.8288	0.9104
No log	2.5	40	0.8210	-0.0556	0.8210	0.9061
No log	2.625	42	0.7835	-0.0727	0.7835	0.8851
No log	2.75	44	0.7786	-0.0514	0.7786	0.8824
No log	2.875	46	0.7565	0.0118	0.7565	0.8698
No log	3.0	48	0.7220	-0.0182	0.7220	0.8497
No log	3.125	50	0.7166	0.0814	0.7166	0.8465
No log	3.25	52	0.7583	0.1360	0.7583	0.8708
No log	3.375	54	0.8157	0.1046	0.8157	0.9032
No log	3.5	56	0.8611	0.1046	0.8611	0.9279
No log	3.625	58	0.7769	0.0885	0.7769	0.8814
No log	3.75	60	0.6909	0.1386	0.6909	0.8312
No log	3.875	62	0.6834	0.2220	0.6834	0.8267
No log	4.0	64	0.7118	0.1764	0.7118	0.8437
No log	4.125	66	0.7684	0.1665	0.7684	0.8766
No log	4.25	68	0.8107	0.0432	0.8107	0.9004
No log	4.375	70	0.7981	0.0932	0.7981	0.8934
No log	4.5	72	0.8419	0.0436	0.8419	0.9175
No log	4.625	74	0.9426	0.0244	0.9426	0.9709
No log	4.75	76	0.9639	-0.0315	0.9639	0.9818
No log	4.875	78	0.9336	0.0297	0.9336	0.9662
No log	5.0	80	0.8545	0.0621	0.8545	0.9244
No log	5.125	82	0.8072	0.0471	0.8072	0.8984
No log	5.25	84	0.8210	0.0611	0.8210	0.9061
No log	5.375	86	0.8904	-0.0145	0.8904	0.9436
No log	5.5	88	0.9487	0.0150	0.9487	0.9740
No log	5.625	90	0.9556	0.0300	0.9556	0.9775
No log	5.75	92	0.9240	-0.0108	0.9240	0.9612
No log	5.875	94	0.8662	-0.0182	0.8662	0.9307
No log	6.0	96	0.7922	0.0185	0.7922	0.8900
No log	6.125	98	0.7830	0.1223	0.7830	0.8849
No log	6.25	100	0.8217	0.0726	0.8217	0.9065
No log	6.375	102	0.8316	0.0887	0.8316	0.9119
No log	6.5	104	0.8117	0.1115	0.8117	0.9009
No log	6.625	106	0.8050	0.1057	0.8050	0.8972
No log	6.75	108	0.8151	0.0634	0.8151	0.9028
No log	6.875	110	0.8651	-0.0294	0.8651	0.9301
No log	7.0	112	0.9607	0.1308	0.9607	0.9802
No log	7.125	114	1.0297	0.1284	1.0297	1.0148
No log	7.25	116	1.0488	0.1403	1.0488	1.0241
No log	7.375	118	1.0257	0.1284	1.0257	1.0128
No log	7.5	120	0.9613	0.1435	0.9613	0.9805
No log	7.625	122	0.8891	0.1569	0.8891	0.9429
No log	7.75	124	0.8229	0.0867	0.8229	0.9071
No log	7.875	126	0.7920	0.1322	0.7920	0.8899
No log	8.0	128	0.7877	0.1372	0.7877	0.8875
No log	8.125	130	0.7929	0.1694	0.7929	0.8905
No log	8.25	132	0.7946	0.1694	0.7946	0.8914
No log	8.375	134	0.7911	0.1372	0.7911	0.8894
No log	8.5	136	0.7913	0.1974	0.7913	0.8895
No log	8.625	138	0.8114	0.1476	0.8114	0.9008
No log	8.75	140	0.8493	0.0728	0.8493	0.9216
No log	8.875	142	0.8907	0.1688	0.8907	0.9437
No log	9.0	144	0.9434	0.1855	0.9434	0.9713
No log	9.125	146	0.9802	0.1553	0.9802	0.9900
No log	9.25	148	1.0073	0.1665	1.0073	1.0036
No log	9.375	150	1.0197	0.1665	1.0197	1.0098
No log	9.5	152	1.0233	0.1665	1.0233	1.0116
No log	9.625	154	1.0233	0.1665	1.0233	1.0116
No log	9.75	156	1.0194	0.1665	1.0194	1.0096
No log	9.875	158	1.0154	0.1665	1.0154	1.0077
No log	10.0	160	1.0147	0.1665	1.0147	1.0073

Framework versions

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