ArabicNewSplits5_FineTuningAraBERT_run1_AugV5_k1_task5_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: 0.7829
• Qwk: 0.6804
• Mse: 0.7829
• Rmse: 0.8848

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	1.8859	0.0519	1.8859	1.3733
No log	0.5	4	1.4316	0.1968	1.4316	1.1965
No log	0.75	6	1.4071	0.1936	1.4071	1.1862
No log	1.0	8	1.3944	0.1663	1.3944	1.1808
No log	1.25	10	1.4234	0.1617	1.4234	1.1931
No log	1.5	12	1.2781	0.2603	1.2781	1.1306
No log	1.75	14	1.1859	0.2916	1.1859	1.0890
No log	2.0	16	1.0914	0.3147	1.0914	1.0447
No log	2.25	18	1.1275	0.3162	1.1275	1.0618
No log	2.5	20	1.1887	0.3171	1.1887	1.0903
No log	2.75	22	1.1925	0.2907	1.1925	1.0920
No log	3.0	24	1.0911	0.3081	1.0911	1.0446
No log	3.25	26	1.0110	0.3392	1.0110	1.0055
No log	3.5	28	0.9762	0.3812	0.9762	0.9880
No log	3.75	30	0.9412	0.4158	0.9412	0.9701
No log	4.0	32	0.9265	0.4646	0.9265	0.9625
No log	4.25	34	0.9662	0.4571	0.9662	0.9830
No log	4.5	36	0.9996	0.4436	0.9996	0.9998
No log	4.75	38	0.9454	0.4623	0.9454	0.9723
No log	5.0	40	0.8604	0.5259	0.8604	0.9276
No log	5.25	42	0.7875	0.6160	0.7875	0.8874
No log	5.5	44	0.7675	0.6212	0.7675	0.8761
No log	5.75	46	0.7682	0.6061	0.7682	0.8764
No log	6.0	48	0.7835	0.6137	0.7835	0.8851
No log	6.25	50	0.7824	0.6212	0.7824	0.8845
No log	6.5	52	0.8241	0.6161	0.8241	0.9078
No log	6.75	54	0.8625	0.6191	0.8625	0.9287
No log	7.0	56	0.8297	0.6393	0.8297	0.9109
No log	7.25	58	0.7828	0.6462	0.7828	0.8848
No log	7.5	60	0.7825	0.6526	0.7825	0.8846
No log	7.75	62	0.7812	0.6563	0.7812	0.8839
No log	8.0	64	0.8024	0.6566	0.8024	0.8958
No log	8.25	66	0.8114	0.6559	0.8114	0.9008
No log	8.5	68	0.8179	0.6559	0.8179	0.9044
No log	8.75	70	0.8225	0.6559	0.8225	0.9069
No log	9.0	72	0.8180	0.6559	0.8180	0.9044
No log	9.25	74	0.8027	0.6559	0.8027	0.8959
No log	9.5	76	0.7906	0.6667	0.7906	0.8892
No log	9.75	78	0.7838	0.6804	0.7838	0.8853
No log	10.0	80	0.7829	0.6804	0.7829	0.8848

Framework versions

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