ArabicNewSplits4_withSameOriginalTrainFileOfSplit3_FineTuningAraBERT_noAug_task3_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: 0.5966
• Qwk: 0.2653
• Mse: 0.5966
• Rmse: 0.7724

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	0.6667	2	3.4268	0.0013	3.4268	1.8512
No log	1.3333	4	2.5942	-0.0732	2.5942	1.6106
No log	2.0	6	1.2332	0.0	1.2332	1.1105
No log	2.6667	8	0.9727	0.0388	0.9727	0.9862
No log	3.3333	10	1.1042	0.0598	1.1042	1.0508
No log	4.0	12	0.7208	0.1590	0.7208	0.8490
No log	4.6667	14	0.6918	0.1590	0.6918	0.8318
No log	5.3333	16	0.6109	0.1913	0.6109	0.7816
No log	6.0	18	0.5634	0.2632	0.5634	0.7506
No log	6.6667	20	0.6051	0.1209	0.6051	0.7779
No log	7.3333	22	0.6485	0.1429	0.6485	0.8053
No log	8.0	24	0.6011	0.2410	0.6011	0.7753
No log	8.6667	26	0.5982	0.2239	0.5982	0.7734
No log	9.3333	28	0.5958	0.2653	0.5958	0.7719
No log	10.0	30	0.5966	0.2653	0.5966	0.7724

Framework versions

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