invoice_extraction_20240816_ana_wrong_aug4x_retrain

This model is a fine-tuned version of naver-clova-ix/donut-base on the imagefolder dataset.

Model description

Trained from Donut base model (naver-clova-ix/donut-base) with Ana v1 model's training data but augmented the ones that got the invoice date wrong even when overfitted.

Intended uses & limitations

More information needed

Training and evaluation data

Train data: 1463 samples

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-06
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

TrainOutput(global_step=14640, training_loss=0.786285320136065, metrics={'train_runtime': 57757.5239, 'train_samples_per_second': 0.507, 'train_steps_per_second': 0.253, 'total_flos': 3.9313585312404455e+19, 'train_loss': 0.786285320136065, 'epoch': 20.0})

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.14.5
• Tokenizers 0.15.2