arabert_baseline_style_task6_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: 0.8646
• Qwk: 0.5642
• Mse: 0.8646

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: 16
• eval_batch_size: 16
• 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
No log	0.5	2	1.9949	0.0653	1.9949
No log	1.0	4	0.9870	0.4463	0.9870
No log	1.5	6	1.0110	0.3212	1.0110
No log	2.0	8	1.2493	0.1888	1.2493
No log	2.5	10	1.3543	0.2877	1.3543
No log	3.0	12	0.9301	0.3348	0.9301
No log	3.5	14	0.8618	0.4615	0.8618
No log	4.0	16	0.8638	0.3275	0.8638
No log	4.5	18	0.8955	0.4340	0.8955
No log	5.0	20	0.9073	0.5391	0.9073
No log	5.5	22	0.8973	0.5817	0.8973
No log	6.0	24	0.9007	0.5817	0.9007
No log	6.5	26	0.8920	0.5817	0.8920
No log	7.0	28	0.8869	0.5817	0.8869
No log	7.5	30	0.8758	0.5817	0.8758
No log	8.0	32	0.8811	0.5642	0.8811
No log	8.5	34	0.8874	0.4943	0.8874
No log	9.0	36	0.8749	0.5642	0.8749
No log	9.5	38	0.8676	0.5642	0.8676
No log	10.0	40	0.8646	0.5642	0.8646

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1