ArabicNewSplits4_WithDuplicationsForScore5_FineTuningAraBERT_run3_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: 0.5792
• Qwk: 0.7568
• Mse: 0.5792
• Rmse: 0.7610

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	2.1249	0.0481	2.1249	1.4577
No log	0.5714	4	1.3339	0.2568	1.3339	1.1549
No log	0.8571	6	1.2672	0.2769	1.2672	1.1257
No log	1.1429	8	1.3756	0.3526	1.3756	1.1728
No log	1.4286	10	1.4595	0.3327	1.4595	1.2081
No log	1.7143	12	1.2101	0.2697	1.2101	1.1000
No log	2.0	14	1.0827	0.3803	1.0827	1.0405
No log	2.2857	16	0.9448	0.4804	0.9448	0.9720
No log	2.5714	18	0.9356	0.5655	0.9356	0.9672
No log	2.8571	20	0.8837	0.6129	0.8837	0.9401
No log	3.1429	22	0.7742	0.6628	0.7742	0.8799
No log	3.4286	24	0.6540	0.6832	0.6540	0.8087
No log	3.7143	26	0.6513	0.6641	0.6513	0.8070
No log	4.0	28	0.6083	0.7228	0.6083	0.7799
No log	4.2857	30	0.5926	0.7369	0.5926	0.7698
No log	4.5714	32	0.6155	0.7467	0.6155	0.7846
No log	4.8571	34	0.5937	0.7554	0.5937	0.7705
No log	5.1429	36	0.5927	0.7228	0.5927	0.7699
No log	5.4286	38	0.6257	0.7036	0.6257	0.7910
No log	5.7143	40	0.5642	0.7654	0.5642	0.7511
No log	6.0	42	0.6725	0.7404	0.6725	0.8201
No log	6.2857	44	0.7244	0.7317	0.7244	0.8511
No log	6.5714	46	0.6215	0.7617	0.6215	0.7883
No log	6.8571	48	0.5527	0.7928	0.5527	0.7435
No log	7.1429	50	0.6085	0.7143	0.6085	0.7801
No log	7.4286	52	0.5957	0.7392	0.5957	0.7718
No log	7.7143	54	0.5433	0.7769	0.5433	0.7371
No log	8.0	56	0.5416	0.7600	0.5416	0.7359
No log	8.2857	58	0.6040	0.7617	0.6040	0.7771
No log	8.5714	60	0.6532	0.7551	0.6532	0.8082
No log	8.8571	62	0.6621	0.7470	0.6621	0.8137
No log	9.1429	64	0.6357	0.7551	0.6357	0.7973
No log	9.4286	66	0.6079	0.7617	0.6079	0.7797
No log	9.7143	68	0.5881	0.7621	0.5881	0.7669
No log	10.0	70	0.5792	0.7568	0.5792	0.7610

Framework versions

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