arabert_no_augmentation_organization_task3_fold0

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: 1.1570
• Qwk: 0.0530
• Mse: 1.1570
• Rmse: 1.0757

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	0.3333	2	5.0689	-0.0476	5.0689	2.2514
No log	0.6667	4	1.7860	-0.0722	1.7860	1.3364
No log	1.0	6	1.1691	-0.0732	1.1691	1.0813
No log	1.3333	8	1.1079	-0.1579	1.1079	1.0526
No log	1.6667	10	1.1020	-0.0565	1.1020	1.0498
No log	2.0	12	1.5769	-0.0942	1.5769	1.2558
No log	2.3333	14	1.4104	0.0	1.4104	1.1876
No log	2.6667	16	1.0770	-0.0565	1.0770	1.0378
No log	3.0	18	0.9699	-0.1159	0.9699	0.9848
No log	3.3333	20	0.9699	-0.3200	0.9699	0.9848
No log	3.6667	22	1.0176	0.2080	1.0176	1.0087
No log	4.0	24	1.0147	0.2080	1.0147	1.0073
No log	4.3333	26	0.9831	-0.1440	0.9831	0.9915
No log	4.6667	28	0.9729	-0.1440	0.9729	0.9864
No log	5.0	30	0.9957	0.0610	0.9957	0.9979
No log	5.3333	32	1.0175	0.0610	1.0175	1.0087
No log	5.6667	34	1.0829	-0.1440	1.0829	1.0406
No log	6.0	36	1.1247	-0.1440	1.1247	1.0605
No log	6.3333	38	1.1230	-0.1440	1.1230	1.0597
No log	6.6667	40	1.1027	0.0435	1.1027	1.0501
No log	7.0	42	1.1089	0.0435	1.1089	1.0530
No log	7.3333	44	1.1313	0.0435	1.1313	1.0636
No log	7.6667	46	1.1576	-0.2384	1.1576	1.0759
No log	8.0	48	1.2109	-0.1440	1.2109	1.1004
No log	8.3333	50	1.2319	-0.1440	1.2319	1.1099
No log	8.6667	52	1.2094	-0.1440	1.2094	1.0997
No log	9.0	54	1.1764	0.0530	1.1764	1.0846
No log	9.3333	56	1.1550	0.0530	1.1550	1.0747
No log	9.6667	58	1.1531	0.0530	1.1531	1.0738
No log	10.0	60	1.1570	0.0530	1.1570	1.0757

Framework versions

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