ArabicNewSplits6_FineTuningAraBERT_run2_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: 1.0685
• Qwk: 0.5510
• Mse: 1.0685
• Rmse: 1.0337

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	2.1859	0.0087	2.1859	1.4785
No log	0.6667	4	1.5282	0.1617	1.5282	1.2362
No log	1.0	6	1.3646	0.1415	1.3646	1.1682
No log	1.3333	8	1.4219	0.3002	1.4219	1.1925
No log	1.6667	10	1.4782	0.3500	1.4782	1.2158
No log	2.0	12	1.5284	0.3253	1.5284	1.2363
No log	2.3333	14	1.6089	0.3389	1.6089	1.2684
No log	2.6667	16	1.5811	0.3774	1.5811	1.2574
No log	3.0	18	1.5889	0.4023	1.5889	1.2605
No log	3.3333	20	1.3708	0.4251	1.3708	1.1708
No log	3.6667	22	1.2599	0.3979	1.2599	1.1224
No log	4.0	24	1.2011	0.4323	1.2011	1.0960
No log	4.3333	26	1.1442	0.4627	1.1442	1.0697
No log	4.6667	28	1.0861	0.5049	1.0861	1.0422
No log	5.0	30	1.0467	0.5192	1.0467	1.0231
No log	5.3333	32	1.0984	0.5116	1.0984	1.0480
No log	5.6667	34	1.1763	0.5190	1.1763	1.0846
No log	6.0	36	1.1968	0.5141	1.1968	1.0940
No log	6.3333	38	1.1624	0.5054	1.1624	1.0782
No log	6.6667	40	1.1057	0.5096	1.1057	1.0515
No log	7.0	42	1.0749	0.5406	1.0749	1.0368
No log	7.3333	44	1.0500	0.5625	1.0500	1.0247
No log	7.6667	46	1.0226	0.5513	1.0226	1.0112
No log	8.0	48	0.9955	0.5583	0.9955	0.9977
No log	8.3333	50	1.0119	0.5709	1.0119	1.0059
No log	8.6667	52	1.0224	0.5520	1.0224	1.0111
No log	9.0	54	1.0394	0.5520	1.0394	1.0195
No log	9.3333	56	1.0515	0.5520	1.0515	1.0254
No log	9.6667	58	1.0645	0.5510	1.0645	1.0317
No log	10.0	60	1.0685	0.5510	1.0685	1.0337

Framework versions

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