ArabicNewSplits3_FineTuningAraBERT_noAug_task1_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: 0.7261
• Qwk: 0.5820
• Mse: 0.7261
• Rmse: 0.8521

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	6.6699	0.0125	6.6699	2.5826
No log	1.0	4	4.7130	0.0884	4.7130	2.1709
No log	1.5	6	2.9566	0.2185	2.9566	1.7195
No log	2.0	8	1.5480	0.1494	1.5480	1.2442
No log	2.5	10	1.1124	0.2500	1.1124	1.0547
No log	3.0	12	1.1315	0.2807	1.1315	1.0637
No log	3.5	14	1.2565	0.2860	1.2565	1.1209
No log	4.0	16	1.1733	0.2260	1.1733	1.0832
No log	4.5	18	1.1568	0.1402	1.1568	1.0755
No log	5.0	20	1.2375	0.2579	1.2375	1.1124
No log	5.5	22	1.1805	0.2848	1.1805	1.0865
No log	6.0	24	1.0408	0.3636	1.0408	1.0202
No log	6.5	26	0.8768	0.5279	0.8768	0.9364
No log	7.0	28	0.7870	0.5356	0.7870	0.8871
No log	7.5	30	0.7522	0.5640	0.7522	0.8673
No log	8.0	32	0.7367	0.5820	0.7367	0.8583
No log	8.5	34	0.7261	0.5820	0.7261	0.8521
No log	9.0	36	0.7232	0.5972	0.7232	0.8504
No log	9.5	38	0.7256	0.5820	0.7256	0.8518
No log	10.0	40	0.7261	0.5820	0.7261	0.8521

Framework versions

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