cor-c's picture
Upload processor
93e8ca7 verified
metadata
license: mit
tags:
  - generated_from_keras_callback
base_model: microsoft/layoutlm-base-uncased
model-index:
  - name: layoutlm-invoice-tf-2
    results: []

layoutlm-invoice-tf-2

This model is a fine-tuned version of microsoft/layoutlm-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

  • Train Loss: 0.3115
  • Validation Loss: 0.3451
  • Train Overall Precision: 0.5831
  • Train Overall Recall: 0.5567
  • Train Overall F1: 0.5696
  • Train Overall Accuracy: 0.8995
  • Epoch: 7

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:

  • optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 3e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
  • training_precision: mixed_float16

Training results

Train Loss Validation Loss Train Overall Precision Train Overall Recall Train Overall F1 Train Overall Accuracy Epoch
2.3655 2.0025 0.0056 0.0101 0.0072 0.3973 0
1.7668 1.5306 0.0185 0.0327 0.0237 0.5455 1
1.3422 1.1230 0.0595 0.0831 0.0693 0.6698 2
1.0006 0.9051 0.1393 0.1990 0.1639 0.7432 3
0.7743 0.6738 0.2127 0.2620 0.2348 0.8158 4
0.5996 0.5262 0.4035 0.4106 0.4070 0.8628 5
0.4179 0.3998 0.4856 0.4685 0.4769 0.8892 6
0.3115 0.3451 0.5831 0.5567 0.5696 0.8995 7

Framework versions

  • Transformers 4.41.0.dev0
  • TensorFlow 2.16.1
  • Datasets 2.19.1
  • Tokenizers 0.19.1