bustamiyusoef/ArabicJawi_NougatBase_v2

ArabicJawi_baseNougat_logs

This model is a fine-tuned version of MohamedRashad/arabic-base-nougat on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0873

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 6
• total_train_batch_size: 48
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss
0.1724	0.9973	187	0.1420
0.1142	2.0	375	0.1124
0.1047	2.9973	562	0.0980
0.0883	4.0	750	0.0895
0.0505	4.9973	937	0.0873
0.0483	6.0	1125	0.0872
0.0387	6.9973	1312	0.0868
0.0261	8.0	1500	0.0861
0.0415	8.9973	1687	0.0891
0.0273	10.0	1875	0.0873

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.0+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3