DANN_JW_OnlyJWAugmentWithJHR

This model is a fine-tuned version of MohamedRashad/arabic-small-nougat on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1478

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss
1.6599	1.0	101	0.3251
1.2198	2.0	202	0.1840
0.9112	3.0	303	0.1631
0.8044	4.0	404	0.1529
0.6899	5.0	505	0.1478
0.7779	6.0	606	0.1473
0.5956	7.0	707	0.1484
0.5914	8.0	808	0.1467
0.5992	9.0	909	0.1440
0.6267	10.0	1010	0.1490
0.5742	11.0	1111	0.1458
0.5247	12.0	1212	0.1477
0.5237	13.0	1313	0.1478
0.5904	14.0	1414	0.1478

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 4.1.0
• Tokenizers 0.21.0