bustamiyusoef/ArabicJawi_NougatBase_v1

_Arabic_base_nougat_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: 0.1325

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
0.2205	0.996	83	0.1900
0.1605	1.992	166	0.1587
0.1008	3.0	250	0.1422
0.1061	3.996	333	0.1317
0.0946	4.992	416	0.1296
0.0619	6.0	500	0.1262
0.0606	6.996	583	0.1279
0.0543	7.992	666	0.1287
0.036	9.0	750	0.1325

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.20.3