lilt-en-funsd / README.md
bathmaraj's picture
End of training
68098f2 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: 1.7110
  • Answer: {'precision': 0.8460661345496009, 'recall': 0.9082007343941249, 'f1': 0.8760330578512396, 'number': 817}
  • Header: {'precision': 0.6470588235294118, 'recall': 0.5546218487394958, 'f1': 0.5972850678733032, 'number': 119}
  • Question: {'precision': 0.9019248395967002, 'recall': 0.9136490250696379, 'f1': 0.9077490774907748, 'number': 1077}
  • Overall Precision: 0.8657
  • Overall Recall: 0.8902
  • Overall F1: 0.8778
  • Overall Accuracy: 0.7988

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
0.4082 10.5263 200 1.0891 {'precision': 0.8286384976525821, 'recall': 0.8641370869033048, 'f1': 0.8460155781905333, 'number': 817} {'precision': 0.4550898203592814, 'recall': 0.6386554621848739, 'f1': 0.5314685314685315, 'number': 119} {'precision': 0.8792134831460674, 'recall': 0.871866295264624, 'f1': 0.8755244755244757, 'number': 1077} 0.8246 0.8549 0.8395 0.7758
0.0489 21.0526 400 1.1864 {'precision': 0.8470588235294118, 'recall': 0.8812729498164015, 'f1': 0.8638272345530894, 'number': 817} {'precision': 0.6078431372549019, 'recall': 0.5210084033613446, 'f1': 0.5610859728506787, 'number': 119} {'precision': 0.8630377524143986, 'recall': 0.9127205199628597, 'f1': 0.8871841155234657, 'number': 1077} 0.8441 0.8768 0.8601 0.8014
0.0135 31.5789 600 1.4159 {'precision': 0.8746928746928747, 'recall': 0.8714810281517748, 'f1': 0.8730839975475169, 'number': 817} {'precision': 0.59375, 'recall': 0.4789915966386555, 'f1': 0.5302325581395348, 'number': 119} {'precision': 0.8701964133219471, 'recall': 0.9461467038068709, 'f1': 0.9065836298932384, 'number': 1077} 0.8592 0.8882 0.8735 0.8040
0.007 42.1053 800 1.4263 {'precision': 0.8548199767711963, 'recall': 0.9008567931456548, 'f1': 0.8772348033373063, 'number': 817} {'precision': 0.6138613861386139, 'recall': 0.5210084033613446, 'f1': 0.5636363636363637, 'number': 119} {'precision': 0.8946412352406903, 'recall': 0.914577530176416, 'f1': 0.9044995408631772, 'number': 1077} 0.8643 0.8857 0.8749 0.8061
0.0039 52.6316 1000 1.6051 {'precision': 0.8764845605700713, 'recall': 0.9033047735618115, 'f1': 0.8896925858951176, 'number': 817} {'precision': 0.5323741007194245, 'recall': 0.6218487394957983, 'f1': 0.5736434108527132, 'number': 119} {'precision': 0.8847209515096066, 'recall': 0.8978644382544104, 'f1': 0.8912442396313364, 'number': 1077} 0.8578 0.8838 0.8706 0.7967
0.0017 63.1579 1200 1.5147 {'precision': 0.8608490566037735, 'recall': 0.8935128518971848, 'f1': 0.8768768768768769, 'number': 817} {'precision': 0.6388888888888888, 'recall': 0.5798319327731093, 'f1': 0.6079295154185022, 'number': 119} {'precision': 0.8934056007226739, 'recall': 0.9182915506035283, 'f1': 0.9056776556776556, 'number': 1077} 0.8667 0.8882 0.8773 0.8087
0.0014 73.6842 1400 1.8128 {'precision': 0.8349514563106796, 'recall': 0.9473684210526315, 'f1': 0.8876146788990826, 'number': 817} {'precision': 0.6078431372549019, 'recall': 0.5210084033613446, 'f1': 0.5610859728506787, 'number': 119} {'precision': 0.9125475285171103, 'recall': 0.8913649025069638, 'f1': 0.9018318459370597, 'number': 1077} 0.8630 0.8922 0.8774 0.7931
0.001 84.2105 1600 1.7309 {'precision': 0.8884758364312267, 'recall': 0.8776009791921665, 'f1': 0.8830049261083744, 'number': 817} {'precision': 0.576271186440678, 'recall': 0.5714285714285714, 'f1': 0.5738396624472574, 'number': 119} {'precision': 0.8825622775800712, 'recall': 0.9210770659238626, 'f1': 0.9014084507042255, 'number': 1077} 0.8673 0.8828 0.8749 0.7998
0.0006 94.7368 1800 1.7644 {'precision': 0.8462414578587699, 'recall': 0.9094247246022031, 'f1': 0.8766961651917403, 'number': 817} {'precision': 0.6363636363636364, 'recall': 0.5294117647058824, 'f1': 0.5779816513761468, 'number': 119} {'precision': 0.9010175763182239, 'recall': 0.904363974001857, 'f1': 0.9026876737720111, 'number': 1077} 0.8649 0.8843 0.8745 0.7967
0.0006 105.2632 2000 1.6953 {'precision': 0.8673835125448028, 'recall': 0.8886168910648715, 'f1': 0.8778718258766626, 'number': 817} {'precision': 0.6666666666666666, 'recall': 0.5378151260504201, 'f1': 0.5953488372093023, 'number': 119} {'precision': 0.8741319444444444, 'recall': 0.9350046425255338, 'f1': 0.9035441902198295, 'number': 1077} 0.8619 0.8927 0.8770 0.8027
0.0003 115.7895 2200 1.7110 {'precision': 0.8460661345496009, 'recall': 0.9082007343941249, 'f1': 0.8760330578512396, 'number': 817} {'precision': 0.6470588235294118, 'recall': 0.5546218487394958, 'f1': 0.5972850678733032, 'number': 119} {'precision': 0.9019248395967002, 'recall': 0.9136490250696379, 'f1': 0.9077490774907748, 'number': 1077} 0.8657 0.8902 0.8778 0.7988
0.0002 126.3158 2400 1.7082 {'precision': 0.8447488584474886, 'recall': 0.9057527539779682, 'f1': 0.874187832250443, 'number': 817} {'precision': 0.6336633663366337, 'recall': 0.5378151260504201, 'f1': 0.5818181818181819, 'number': 119} {'precision': 0.9002744739249772, 'recall': 0.9136490250696379, 'f1': 0.9069124423963134, 'number': 1077} 0.8638 0.8882 0.8758 0.7978

Framework versions

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