DANN_CH_2

This model is a fine-tuned version of bustamiyusoef/Nougat_CH_20k_randomly on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5035

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
5.6456	1.0	100	0.9039
4.1682	2.0	200	0.7489
3.1162	3.0	300	0.5641
1.9667	4.0	400	0.5789
1.6862	5.0	500	0.7187
1.2244	6.0	600	0.5561
1.2133	7.0	700	0.5199
0.9258	8.0	800	0.5285
0.8586	9.0	900	0.4849
0.6571	10.0	1000	0.4835
0.6191	11.0	1100	0.4929
0.5845	12.0	1200	0.4899
0.5797	13.0	1300	0.4943
0.6118	14.0	1400	0.5066
0.5977	15.0	1500	0.5035

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 4.1.1
• Tokenizers 0.21.0