ArabicNewSplits4_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: 0.5618
• Qwk: 0.7601
• Mse: 0.5618
• Rmse: 0.7495

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	2.0819	0.0345	2.0819	1.4429
No log	0.5714	4	1.3364	0.2722	1.3364	1.1560
No log	0.8571	6	1.1764	0.3108	1.1764	1.0846
No log	1.1429	8	1.1814	0.3614	1.1814	1.0869
No log	1.4286	10	1.0261	0.4398	1.0261	1.0130
No log	1.7143	12	1.3281	0.4409	1.3281	1.1524
No log	2.0	14	1.4901	0.4138	1.4901	1.2207
No log	2.2857	16	1.3984	0.4610	1.3984	1.1825
No log	2.5714	18	1.0130	0.5398	1.0130	1.0065
No log	2.8571	20	0.7322	0.6297	0.7322	0.8557
No log	3.1429	22	0.7589	0.5922	0.7589	0.8711
No log	3.4286	24	0.6645	0.6208	0.6645	0.8151
No log	3.7143	26	0.6739	0.6826	0.6739	0.8209
No log	4.0	28	0.9998	0.6224	0.9998	0.9999
No log	4.2857	30	1.1582	0.5483	1.1582	1.0762
No log	4.5714	32	1.1064	0.5585	1.1064	1.0519
No log	4.8571	34	0.9032	0.6698	0.9032	0.9504
No log	5.1429	36	0.7053	0.6930	0.7053	0.8398
No log	5.4286	38	0.6157	0.7218	0.6157	0.7846
No log	5.7143	40	0.6153	0.7218	0.6153	0.7844
No log	6.0	42	0.6056	0.7218	0.6056	0.7782
No log	6.2857	44	0.5847	0.7243	0.5847	0.7646
No log	6.5714	46	0.5866	0.7410	0.5866	0.7659
No log	6.8571	48	0.6453	0.7109	0.6453	0.8033
No log	7.1429	50	0.7782	0.7044	0.7782	0.8822
No log	7.4286	52	0.8748	0.7119	0.8748	0.9353
No log	7.7143	54	0.8713	0.7087	0.8713	0.9334
No log	8.0	56	0.8172	0.7130	0.8172	0.9040
No log	8.2857	58	0.7465	0.7130	0.7465	0.8640
No log	8.5714	60	0.6614	0.7195	0.6614	0.8133
No log	8.8571	62	0.5955	0.7432	0.5955	0.7717
No log	9.1429	64	0.5714	0.7727	0.5714	0.7559
No log	9.4286	66	0.5617	0.7601	0.5617	0.7495
No log	9.7143	68	0.5608	0.7601	0.5608	0.7489
No log	10.0	70	0.5618	0.7601	0.5618	0.7495

Framework versions

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