_Arabic_nougat_AHRXjawi

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: 0.2568

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss
1.948	0.9969	159	0.2862
1.2901	1.9969	318	0.2339
1.1543	2.9969	477	0.2213
0.8668	3.9969	636	0.2139
0.6597	4.9969	795	0.2151
0.6014	5.9969	954	0.2153
0.418	6.9969	1113	0.2250
0.3342	7.9969	1272	0.2305
0.2786	8.9969	1431	0.2333
0.2538	9.9969	1590	0.2419
0.1871	10.9969	1749	0.2399
0.2311	11.9969	1908	0.2467
0.1756	12.9969	2067	0.2523
0.1788	13.9969	2226	0.2528
0.162	14.9969	2385	0.2546
0.1526	15.9969	2544	0.2558
0.1331	16.9969	2703	0.2568
0.1614	17.9969	2862	0.2568
0.1636	18.9969	3021	0.2571
0.1422	19.9969	3180	0.2568

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0