ArabicNewSplits4_withSameOriginalTrainFileOfSplit3_FineTuningAraBERT_noAug_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: 0.7735
• Qwk: 0.4301
• Mse: 0.7735
• Rmse: 0.8795

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	0.5	2	3.9949	0.0352	3.9949	1.9987
No log	1.0	4	2.1097	0.0566	2.1097	1.4525
No log	1.5	6	1.0984	0.0864	1.0984	1.0480
No log	2.0	8	0.8102	0.0191	0.8102	0.9001
No log	2.5	10	0.7324	0.1841	0.7324	0.8558
No log	3.0	12	0.7691	0.1585	0.7691	0.8770
No log	3.5	14	0.9091	0.1073	0.9091	0.9535
No log	4.0	16	0.8613	0.2259	0.8613	0.9280
No log	4.5	18	0.7906	0.2998	0.7906	0.8892
No log	5.0	20	0.8261	0.3354	0.8261	0.9089
No log	5.5	22	1.0001	0.2999	1.0001	1.0001
No log	6.0	24	0.9977	0.2453	0.9977	0.9989
No log	6.5	26	0.9361	0.2159	0.9361	0.9675
No log	7.0	28	0.8738	0.3377	0.8738	0.9348
No log	7.5	30	0.8251	0.3290	0.8251	0.9083
No log	8.0	32	0.7739	0.4360	0.7739	0.8797
No log	8.5	34	0.7416	0.4903	0.7416	0.8612
No log	9.0	36	0.7446	0.4757	0.7446	0.8629
No log	9.5	38	0.7633	0.4459	0.7633	0.8737
No log	10.0	40	0.7735	0.4301	0.7735	0.8795

Framework versions

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