ArabicNewSplits_FineTuningAraBERT_AugV5_k1_task2_organization_fold1

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: 1.2144
• Qwk: 0.1467
• Mse: 1.2144
• Rmse: 1.1020

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	0.4	2	3.7271	-0.0605	3.7271	1.9306
No log	0.8	4	1.5854	0.1468	1.5854	1.2591
No log	1.2	6	0.7302	0.1002	0.7302	0.8545
No log	1.6	8	0.6196	0.3564	0.6196	0.7872
No log	2.0	10	0.8956	-0.2529	0.8956	0.9463
No log	2.4	12	1.0018	-0.0601	1.0018	1.0009
No log	2.8	14	0.8659	-0.1060	0.8659	0.9305
No log	3.2	16	0.8088	-0.0357	0.8088	0.8993
No log	3.6	18	0.7840	-0.0038	0.7840	0.8855
No log	4.0	20	0.7839	0.0025	0.7839	0.8854
No log	4.4	22	0.8070	-0.0650	0.8070	0.8983
No log	4.8	24	0.8483	-0.0264	0.8483	0.9210
No log	5.2	26	0.8709	-0.0264	0.8709	0.9332
No log	5.6	28	0.8727	-0.0650	0.8727	0.9342
No log	6.0	30	0.8935	-0.0533	0.8935	0.9452
No log	6.4	32	0.9479	-0.0108	0.9479	0.9736
No log	6.8	34	1.0149	-0.0264	1.0149	1.0074
No log	7.2	36	1.0203	-0.0264	1.0203	1.0101
No log	7.6	38	1.0270	0.0582	1.0270	1.0134
No log	8.0	40	1.0451	0.1273	1.0451	1.0223
No log	8.4	42	1.0897	0.0985	1.0897	1.0439
No log	8.8	44	1.1484	0.1253	1.1484	1.0717
No log	9.2	46	1.1819	0.1467	1.1819	1.0871
No log	9.6	48	1.2013	0.1467	1.2013	1.0960
No log	10.0	50	1.2144	0.1467	1.2144	1.1020

Framework versions

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