ArabicNewSplits4_WithDuplicationsForScore5_FineTuningAraBERT_run2_AugV5_k1_task2_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: 0.8311
• Qwk: 0.5422
• Mse: 0.8311
• Rmse: 0.9116

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	0.2222	2	3.9392	-0.0062	3.9392	1.9847
No log	0.4444	4	2.2132	0.0779	2.2132	1.4877
No log	0.6667	6	1.0749	0.0666	1.0749	1.0368
No log	0.8889	8	0.8034	0.0360	0.8034	0.8963
No log	1.1111	10	0.7452	0.0993	0.7452	0.8632
No log	1.3333	12	0.7030	0.1834	0.7030	0.8385
No log	1.5556	14	0.6977	0.2149	0.6977	0.8353
No log	1.7778	16	0.6906	0.2236	0.6906	0.8310
No log	2.0	18	0.7289	0.2375	0.7289	0.8538
No log	2.2222	20	1.2202	0.1106	1.2202	1.1046
No log	2.4444	22	1.2447	0.1315	1.2447	1.1157
No log	2.6667	24	0.7872	0.3002	0.7872	0.8872
No log	2.8889	26	0.6410	0.3193	0.6410	0.8006
No log	3.1111	28	0.7345	0.3653	0.7345	0.8570
No log	3.3333	30	0.7517	0.3996	0.7517	0.8670
No log	3.5556	32	0.6636	0.4458	0.6636	0.8146
No log	3.7778	34	0.6848	0.3956	0.6848	0.8275
No log	4.0	36	0.7934	0.3506	0.7934	0.8907
No log	4.2222	38	0.7878	0.3674	0.7878	0.8876
No log	4.4444	40	0.6736	0.4128	0.6736	0.8208
No log	4.6667	42	0.6528	0.4250	0.6528	0.8079
No log	4.8889	44	0.8012	0.4283	0.8012	0.8951
No log	5.1111	46	0.8557	0.3546	0.8557	0.9251
No log	5.3333	48	0.7751	0.4221	0.7751	0.8804
No log	5.5556	50	0.6745	0.4812	0.6745	0.8213
No log	5.7778	52	0.6712	0.4691	0.6712	0.8192
No log	6.0	54	0.7760	0.4835	0.7760	0.8809
No log	6.2222	56	0.8589	0.4930	0.8589	0.9268
No log	6.4444	58	0.8488	0.5044	0.8488	0.9213
No log	6.6667	60	0.7760	0.4822	0.7760	0.8809
No log	6.8889	62	0.7641	0.4928	0.7641	0.8741
No log	7.1111	64	0.7725	0.4986	0.7725	0.8789
No log	7.3333	66	0.7931	0.5133	0.7931	0.8906
No log	7.5556	68	0.8187	0.5229	0.8187	0.9048
No log	7.7778	70	0.8212	0.5316	0.8212	0.9062
No log	8.0	72	0.8327	0.5106	0.8327	0.9125
No log	8.2222	74	0.8441	0.5054	0.8441	0.9188
No log	8.4444	76	0.8551	0.4944	0.8551	0.9247
No log	8.6667	78	0.8512	0.5054	0.8512	0.9226
No log	8.8889	80	0.8390	0.5106	0.8390	0.9160
No log	9.1111	82	0.8315	0.5270	0.8315	0.9119
No log	9.3333	84	0.8348	0.5274	0.8348	0.9137
No log	9.5556	86	0.8345	0.5270	0.8345	0.9135
No log	9.7778	88	0.8325	0.5424	0.8325	0.9124
No log	10.0	90	0.8311	0.5422	0.8311	0.9116

Framework versions

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