donut_experiment_bayesian_trial_13

This model is a fine-tuned version of naver-clova-ix/donut-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5500
• Bleu: 0.0682
• Precisions: [0.8219461697722568, 0.755868544600939, 0.7073170731707317, 0.6474358974358975]
• Brevity Penalty: 0.0934
• Length Ratio: 0.2967
• Translation Length: 483
• Reference Length: 1628
• Cer: 0.7531
• Wer: 0.8285

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: 6.0814226870239416e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 2
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Precisions	Brevity Penalty	Length Ratio	Translation Length	Reference Length	Cer	Wer
0.0455	1.0	253	0.5262	0.0660	[0.8166666666666667, 0.7446808510638298, 0.6939890710382514, 0.6407766990291263]	0.0915	0.2948	480	1628	0.7601	0.8316
0.0276	2.0	506	0.5500	0.0682	[0.8219461697722568, 0.755868544600939, 0.7073170731707317, 0.6474358974358975]	0.0934	0.2967	483	1628	0.7531	0.8285

Framework versions

• Transformers 4.40.0
• Pytorch 2.1.0
• Datasets 2.18.0
• Tokenizers 0.19.1