ArabicNewSplits6_WithDuplicationsForScore5_FineTuningAraBERT_run3_AugV5_k1_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: 0.8663
• Qwk: 0.5278
• Mse: 0.8663
• Rmse: 0.9307

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	0.25	2	3.7307	0.0140	3.7307	1.9315
No log	0.5	4	1.9939	0.0717	1.9939	1.4121
No log	0.75	6	1.0296	0.0290	1.0296	1.0147
No log	1.0	8	0.7836	-0.0086	0.7836	0.8852
No log	1.25	10	0.7590	0.2057	0.7590	0.8712
No log	1.5	12	0.8124	0.2101	0.8124	0.9013
No log	1.75	14	0.8359	0.2181	0.8359	0.9143
No log	2.0	16	0.8082	0.1601	0.8082	0.8990
No log	2.25	18	0.7767	0.2326	0.7767	0.8813
No log	2.5	20	0.8741	0.2302	0.8741	0.9349
No log	2.75	22	0.9425	0.2644	0.9425	0.9708
No log	3.0	24	0.8150	0.3077	0.8150	0.9028
No log	3.25	26	0.6937	0.3019	0.6937	0.8329
No log	3.5	28	0.6478	0.3947	0.6478	0.8049
No log	3.75	30	0.6743	0.3460	0.6743	0.8212
No log	4.0	32	0.6837	0.3607	0.6837	0.8269
No log	4.25	34	0.6907	0.3732	0.6907	0.8311
No log	4.5	36	0.7052	0.3906	0.7052	0.8398
No log	4.75	38	0.7075	0.4158	0.7075	0.8411
No log	5.0	40	0.7488	0.4586	0.7488	0.8653
No log	5.25	42	0.7741	0.4602	0.7741	0.8798
No log	5.5	44	0.7378	0.4632	0.7378	0.8590
No log	5.75	46	0.7285	0.4773	0.7285	0.8535
No log	6.0	48	0.7405	0.4910	0.7405	0.8605
No log	6.25	50	0.7991	0.4946	0.7991	0.8939
No log	6.5	52	0.7877	0.4744	0.7877	0.8875
No log	6.75	54	0.8212	0.4915	0.8212	0.9062
No log	7.0	56	0.8341	0.4915	0.8341	0.9133
No log	7.25	58	0.7945	0.5016	0.7945	0.8914
No log	7.5	60	0.8125	0.4961	0.8125	0.9014
No log	7.75	62	0.8046	0.5108	0.8046	0.8970
No log	8.0	64	0.8334	0.5288	0.8334	0.9129
No log	8.25	66	0.8299	0.5365	0.8299	0.9110
No log	8.5	68	0.8204	0.5158	0.8204	0.9058
No log	8.75	70	0.8294	0.5196	0.8294	0.9107
No log	9.0	72	0.8340	0.5196	0.8340	0.9132
No log	9.25	74	0.8547	0.5405	0.8547	0.9245
No log	9.5	76	0.8685	0.5278	0.8685	0.9320
No log	9.75	78	0.8683	0.5278	0.8683	0.9318
No log	10.0	80	0.8663	0.5278	0.8663	0.9307

Framework versions

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