Nougat_GER_x_EHR

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

• Loss: 0.1708

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: 10
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 6
• total_train_batch_size: 60
• 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
1.5079	1.0	67	0.2175
0.7336	2.0	134	0.1942
0.2775	3.0	201	0.1904
0.1441	4.0	268	0.1825
0.0805	5.0	335	0.1798
0.0517	6.0	402	0.1795
0.0451	7.0	469	0.1771
0.0284	8.0	536	0.1743
0.0182	9.0	603	0.1710
0.0143	10.0	670	0.1725
0.0105	11.0	737	0.1713
0.013	12.0	804	0.1712
0.0096	13.0	871	0.1708
0.0087	14.0	938	0.1707
0.0093	14.78	990	0.1708

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 4.0.0
• Tokenizers 0.21.0