ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run3_AugV5_k1_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: 0.9605
• Qwk: 0.1148
• Mse: 0.9605
• Rmse: 0.9801

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	0.2857	2	3.1351	-0.0138	3.1351	1.7706
No log	0.5714	4	1.6487	-0.0070	1.6487	1.2840
No log	0.8571	6	0.9428	0.0418	0.9428	0.9710
No log	1.1429	8	1.0210	0.0847	1.0210	1.0104
No log	1.4286	10	0.6218	0.0476	0.6218	0.7886
No log	1.7143	12	0.7490	0.125	0.7490	0.8655
No log	2.0	14	0.7612	0.0	0.7612	0.8725
No log	2.2857	16	0.6580	0.0	0.6580	0.8112
No log	2.5714	18	0.5776	0.0569	0.5776	0.7600
No log	2.8571	20	0.6125	-0.0370	0.6125	0.7826
No log	3.1429	22	0.6701	0.0189	0.6701	0.8186
No log	3.4286	24	0.6421	-0.0133	0.6421	0.8013
No log	3.7143	26	0.6429	-0.0963	0.6429	0.8018
No log	4.0	28	0.6664	0.1176	0.6664	0.8164
No log	4.2857	30	0.7573	0.2318	0.7573	0.8702
No log	4.5714	32	0.7705	0.1795	0.7705	0.8778
No log	4.8571	34	0.7629	0.0233	0.7629	0.8735
No log	5.1429	36	0.8292	0.0455	0.8292	0.9106
No log	5.4286	38	0.8067	0.0497	0.8067	0.8982
No log	5.7143	40	0.8222	0.1304	0.8222	0.9068
No log	6.0	42	0.8578	0.1131	0.8578	0.9262
No log	6.2857	44	0.8520	0.0980	0.8520	0.9230
No log	6.5714	46	0.8675	0.0857	0.8675	0.9314
No log	6.8571	48	0.8601	0.0741	0.8601	0.9274
No log	7.1429	50	0.8865	0.1131	0.8865	0.9416
No log	7.4286	52	0.9160	0.1203	0.9160	0.9571
No log	7.7143	54	0.9153	0.1186	0.9153	0.9567
No log	8.0	56	0.8859	0.1660	0.8859	0.9412
No log	8.2857	58	0.8600	0.2212	0.8600	0.9274
No log	8.5714	60	0.8610	0.2070	0.8610	0.9279
No log	8.8571	62	0.8818	0.1864	0.8818	0.9390
No log	9.1429	64	0.9196	0.1074	0.9196	0.9589
No log	9.4286	66	0.9461	0.1148	0.9461	0.9727
No log	9.7143	68	0.9583	0.1148	0.9583	0.9789
No log	10.0	70	0.9605	0.1148	0.9605	0.9801

Framework versions

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