lilt-en-funsd / README.md
indudesane's picture
End of training
72c0978 verified
metadata
license: mit
base_model: SCUT-DLVCLab/lilt-roberta-en-base
tags:
  - generated_from_trainer
model-index:
  - name: lilt-en-funsd
    results: []

lilt-en-funsd

This model is a fine-tuned version of SCUT-DLVCLab/lilt-roberta-en-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.5687
  • Answer: {'precision': 0.8635294117647059, 'recall': 0.8984088127294981, 'f1': 0.8806238752249551, 'number': 817}
  • Header: {'precision': 0.6781609195402298, 'recall': 0.4957983193277311, 'f1': 0.5728155339805825, 'number': 119}
  • Question: {'precision': 0.8912655971479501, 'recall': 0.9285051067780873, 'f1': 0.9095043201455207, 'number': 1077}
  • Overall Precision: 0.8708
  • Overall Recall: 0.8907
  • Overall F1: 0.8806
  • Overall Accuracy: 0.8098

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: 5e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • training_steps: 2500
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Answer Header Question Overall Precision Overall Recall Overall F1 Overall Accuracy
0.4119 10.5263 200 1.0913 {'precision': 0.8155555555555556, 'recall': 0.8984088127294981, 'f1': 0.8549796156086196, 'number': 817} {'precision': 0.5523809523809524, 'recall': 0.48739495798319327, 'f1': 0.5178571428571428, 'number': 119} {'precision': 0.8904363974001857, 'recall': 0.8904363974001857, 'f1': 0.8904363974001857, 'number': 1077} 0.8410 0.8698 0.8552 0.7742
0.0512 21.0526 400 1.3167 {'precision': 0.8363636363636363, 'recall': 0.9008567931456548, 'f1': 0.8674130819092517, 'number': 817} {'precision': 0.6160714285714286, 'recall': 0.5798319327731093, 'f1': 0.5974025974025975, 'number': 119} {'precision': 0.8839122486288848, 'recall': 0.8978644382544104, 'f1': 0.8908337171810224, 'number': 1077} 0.8495 0.8803 0.8646 0.7919
0.0159 31.5789 600 1.4418 {'precision': 0.8539458186101295, 'recall': 0.8873929008567931, 'f1': 0.8703481392557022, 'number': 817} {'precision': 0.5913978494623656, 'recall': 0.46218487394957986, 'f1': 0.5188679245283019, 'number': 119} {'precision': 0.8617113223854796, 'recall': 0.9257195914577531, 'f1': 0.8925693822739481, 'number': 1077} 0.8466 0.8828 0.8643 0.7994
0.0085 42.1053 800 1.3252 {'precision': 0.8408839779005525, 'recall': 0.9314565483476133, 'f1': 0.8838559814169571, 'number': 817} {'precision': 0.6176470588235294, 'recall': 0.5294117647058824, 'f1': 0.5701357466063349, 'number': 119} {'precision': 0.8946412352406903, 'recall': 0.914577530176416, 'f1': 0.9044995408631772, 'number': 1077} 0.8582 0.8987 0.8779 0.8077
0.0035 52.6316 1000 1.4334 {'precision': 0.8364269141531323, 'recall': 0.8824969400244798, 'f1': 0.8588445503275759, 'number': 817} {'precision': 0.6, 'recall': 0.5798319327731093, 'f1': 0.5897435897435898, 'number': 119} {'precision': 0.8983516483516484, 'recall': 0.9108635097493036, 'f1': 0.9045643153526972, 'number': 1077} 0.8560 0.8798 0.8677 0.7988
0.0031 63.1579 1200 1.3464 {'precision': 0.8569780853517878, 'recall': 0.9094247246022031, 'f1': 0.8824228028503563, 'number': 817} {'precision': 0.6836734693877551, 'recall': 0.5630252100840336, 'f1': 0.6175115207373272, 'number': 119} {'precision': 0.8909090909090909, 'recall': 0.9099350046425255, 'f1': 0.90032154340836, 'number': 1077} 0.8668 0.8892 0.8779 0.8164
0.0017 73.6842 1400 1.5019 {'precision': 0.883054892601432, 'recall': 0.9057527539779682, 'f1': 0.8942598187311178, 'number': 817} {'precision': 0.6565656565656566, 'recall': 0.5462184873949579, 'f1': 0.5963302752293578, 'number': 119} {'precision': 0.8922528940338379, 'recall': 0.9303621169916435, 'f1': 0.9109090909090909, 'number': 1077} 0.8772 0.8977 0.8873 0.8110
0.0008 84.2105 1600 1.5955 {'precision': 0.8760631834750912, 'recall': 0.8824969400244798, 'f1': 0.8792682926829267, 'number': 817} {'precision': 0.5925925925925926, 'recall': 0.5378151260504201, 'f1': 0.5638766519823789, 'number': 119} {'precision': 0.8853046594982079, 'recall': 0.9173630454967502, 'f1': 0.9010487916096672, 'number': 1077} 0.8661 0.8808 0.8734 0.8084
0.0006 94.7368 1800 1.5931 {'precision': 0.8713592233009708, 'recall': 0.8788249694002448, 'f1': 0.8750761730652041, 'number': 817} {'precision': 0.6122448979591837, 'recall': 0.5042016806722689, 'f1': 0.5529953917050692, 'number': 119} {'precision': 0.8834519572953736, 'recall': 0.9220055710306406, 'f1': 0.9023171285779192, 'number': 1077} 0.8656 0.8798 0.8726 0.7973
0.0006 105.2632 2000 1.5735 {'precision': 0.8676122931442081, 'recall': 0.8984088127294981, 'f1': 0.8827420324714371, 'number': 817} {'precision': 0.6176470588235294, 'recall': 0.5294117647058824, 'f1': 0.5701357466063349, 'number': 119} {'precision': 0.8862222222222222, 'recall': 0.9257195914577531, 'f1': 0.9055404178019981, 'number': 1077} 0.8654 0.8912 0.8781 0.8002
0.001 115.7895 2200 1.5605 {'precision': 0.8543352601156069, 'recall': 0.9045287637698899, 'f1': 0.8787158145065399, 'number': 817} {'precision': 0.6559139784946236, 'recall': 0.5126050420168067, 'f1': 0.5754716981132076, 'number': 119} {'precision': 0.8935978358881875, 'recall': 0.9201485608170845, 'f1': 0.9066788655077767, 'number': 1077} 0.8665 0.8897 0.8779 0.8073
0.0004 126.3158 2400 1.5687 {'precision': 0.8635294117647059, 'recall': 0.8984088127294981, 'f1': 0.8806238752249551, 'number': 817} {'precision': 0.6781609195402298, 'recall': 0.4957983193277311, 'f1': 0.5728155339805825, 'number': 119} {'precision': 0.8912655971479501, 'recall': 0.9285051067780873, 'f1': 0.9095043201455207, 'number': 1077} 0.8708 0.8907 0.8806 0.8098

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.2
  • Tokenizers 0.19.1