Edit model card

layoutlmv3-finetuned-cord_100

This model is a fine-tuned version of microsoft/layoutlmv3-base on the cord-layoutlmv3 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4548
  • Precision: 0.8760
  • Recall: 0.8983
  • F1: 0.8870
  • Accuracy: 0.9147

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: 1e-05
  • train_batch_size: 5
  • eval_batch_size: 5
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • training_steps: 2500

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 31.25 250 0.7175 0.7874 0.8475 0.8163 0.8768
1.0535 62.5 500 0.4279 0.8843 0.9068 0.8954 0.9147
1.0535 93.75 750 0.4042 0.8760 0.8983 0.8870 0.9147
0.0727 125.0 1000 0.4065 0.8760 0.8983 0.8870 0.9194
0.0727 156.25 1250 0.4290 0.8843 0.9068 0.8954 0.9147
0.0245 187.5 1500 0.4511 0.8760 0.8983 0.8870 0.9100
0.0245 218.75 1750 0.4594 0.8760 0.8983 0.8870 0.9147
0.0155 250.0 2000 0.4566 0.8760 0.8983 0.8870 0.9147
0.0155 281.25 2250 0.4489 0.8760 0.8983 0.8870 0.9147
0.0125 312.5 2500 0.4548 0.8760 0.8983 0.8870 0.9147

Framework versions

  • Transformers 4.28.0
  • Pytorch 2.0.1+cu118
  • Datasets 2.14.4
  • Tokenizers 0.13.3
Downloads last month
2

Evaluation results