layoutlm-funsd / README.md
sreejith8100's picture
End of training
1ac99ba
metadata
base_model: microsoft/layoutlm-base-uncased
tags:
  - generated_from_trainer
datasets:
  - funsd
model-index:
  - name: layoutlm-funsd
    results: []

layoutlm-funsd

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

  • Loss: 0.7204
  • Answer: {'precision': 0.7103218645948945, 'recall': 0.7911001236093943, 'f1': 0.7485380116959064, 'number': 809}
  • Header: {'precision': 0.3697478991596639, 'recall': 0.3697478991596639, 'f1': 0.3697478991596639, 'number': 119}
  • Question: {'precision': 0.7799126637554585, 'recall': 0.8384976525821596, 'f1': 0.8081447963800905, 'number': 1065}
  • Overall Precision: 0.7284
  • Overall Recall: 0.7913
  • Overall F1: 0.7585
  • Overall Accuracy: 0.7930

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: 3e-05
  • train_batch_size: 16
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 15

Training results

Training Loss Epoch Step Validation Loss Answer Header Question Overall Precision Overall Recall Overall F1 Overall Accuracy
1.8258 1.0 10 1.6104 {'precision': 0.03933136676499508, 'recall': 0.049443757725587144, 'f1': 0.04381161007667032, 'number': 809} {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} {'precision': 0.2102076124567474, 'recall': 0.22816901408450704, 'f1': 0.21882035119315627, 'number': 1065} 0.1302 0.1420 0.1359 0.3742
1.4584 2.0 20 1.2730 {'precision': 0.1923509561304837, 'recall': 0.21137206427688504, 'f1': 0.20141342756183747, 'number': 809} {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} {'precision': 0.4099290780141844, 'recall': 0.5427230046948357, 'f1': 0.467070707070707, 'number': 1065} 0.3258 0.3758 0.3490 0.5734
1.1076 3.0 30 0.9718 {'precision': 0.501532175689479, 'recall': 0.6069221260815822, 'f1': 0.5492170022371365, 'number': 809} {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} {'precision': 0.5749211356466877, 'recall': 0.6845070422535211, 'f1': 0.6249464209172738, 'number': 1065} 0.5358 0.6121 0.5714 0.6890
0.8541 4.0 40 0.8287 {'precision': 0.5573921028466483, 'recall': 0.7503090234857849, 'f1': 0.6396206533192834, 'number': 809} {'precision': 0.14893617021276595, 'recall': 0.058823529411764705, 'f1': 0.08433734939759036, 'number': 119} {'precision': 0.6451342281879194, 'recall': 0.7220657276995305, 'f1': 0.6814355338945502, 'number': 1065} 0.5941 0.6939 0.6401 0.7420
0.7105 5.0 50 0.7483 {'precision': 0.6358695652173914, 'recall': 0.723114956736712, 'f1': 0.6766917293233082, 'number': 809} {'precision': 0.2465753424657534, 'recall': 0.15126050420168066, 'f1': 0.18749999999999997, 'number': 119} {'precision': 0.6898305084745763, 'recall': 0.7643192488262911, 'f1': 0.7251670378619154, 'number': 1065} 0.6521 0.7110 0.6803 0.7618
0.6079 6.0 60 0.7023 {'precision': 0.6306209850107066, 'recall': 0.7280593325092707, 'f1': 0.6758462421113023, 'number': 809} {'precision': 0.2875, 'recall': 0.19327731092436976, 'f1': 0.23115577889447236, 'number': 119} {'precision': 0.6796267496111975, 'recall': 0.8206572769953052, 'f1': 0.7435133985538068, 'number': 1065} 0.6461 0.7456 0.6923 0.7776
0.5267 7.0 70 0.6779 {'precision': 0.674892703862661, 'recall': 0.7775030902348579, 'f1': 0.7225732337736933, 'number': 809} {'precision': 0.3, 'recall': 0.226890756302521, 'f1': 0.25837320574162675, 'number': 119} {'precision': 0.717391304347826, 'recall': 0.8056338028169014, 'f1': 0.7589562140645733, 'number': 1065} 0.6826 0.7597 0.7191 0.7853
0.4735 8.0 80 0.6688 {'precision': 0.6955093099671413, 'recall': 0.7849196538936959, 'f1': 0.7375145180023228, 'number': 809} {'precision': 0.32608695652173914, 'recall': 0.25210084033613445, 'f1': 0.2843601895734597, 'number': 119} {'precision': 0.7424892703862661, 'recall': 0.812206572769953, 'f1': 0.7757847533632287, 'number': 1065} 0.7051 0.7677 0.7350 0.7950
0.4196 9.0 90 0.6791 {'precision': 0.6843243243243243, 'recall': 0.7824474660074165, 'f1': 0.7301038062283737, 'number': 809} {'precision': 0.3181818181818182, 'recall': 0.29411764705882354, 'f1': 0.3056768558951965, 'number': 119} {'precision': 0.7561807331628303, 'recall': 0.8328638497652582, 'f1': 0.7926720285969615, 'number': 1065} 0.7043 0.7802 0.7403 0.7937
0.3756 10.0 100 0.6968 {'precision': 0.7089887640449438, 'recall': 0.7799752781211372, 'f1': 0.7427898763978811, 'number': 809} {'precision': 0.328, 'recall': 0.3445378151260504, 'f1': 0.33606557377049184, 'number': 119} {'precision': 0.7790393013100436, 'recall': 0.8375586854460094, 'f1': 0.807239819004525, 'number': 1065} 0.7241 0.7847 0.7532 0.7947
0.3402 11.0 110 0.6959 {'precision': 0.7024070021881839, 'recall': 0.7935723114956736, 'f1': 0.7452118398142775, 'number': 809} {'precision': 0.3416666666666667, 'recall': 0.3445378151260504, 'f1': 0.34309623430962344, 'number': 119} {'precision': 0.7791411042944786, 'recall': 0.8347417840375587, 'f1': 0.8059836808703537, 'number': 1065} 0.7228 0.7888 0.7543 0.7958
0.3225 12.0 120 0.6945 {'precision': 0.7106430155210643, 'recall': 0.792336217552534, 'f1': 0.7492694330800702, 'number': 809} {'precision': 0.3644067796610169, 'recall': 0.36134453781512604, 'f1': 0.3628691983122363, 'number': 119} {'precision': 0.7667238421955404, 'recall': 0.8394366197183099, 'f1': 0.8014343343792021, 'number': 1065} 0.7219 0.7918 0.7552 0.7961
0.3031 13.0 130 0.7204 {'precision': 0.71, 'recall': 0.7898640296662547, 'f1': 0.7478057343475717, 'number': 809} {'precision': 0.35, 'recall': 0.35294117647058826, 'f1': 0.35146443514644354, 'number': 119} {'precision': 0.7895204262877442, 'recall': 0.8347417840375587, 'f1': 0.8115015974440895, 'number': 1065} 0.7316 0.7878 0.7586 0.7916
0.289 14.0 140 0.7196 {'precision': 0.7095709570957096, 'recall': 0.7972805933250927, 'f1': 0.750873108265425, 'number': 809} {'precision': 0.3826086956521739, 'recall': 0.3697478991596639, 'f1': 0.37606837606837606, 'number': 119} {'precision': 0.7816593886462883, 'recall': 0.8403755868544601, 'f1': 0.8099547511312217, 'number': 1065} 0.7303 0.7948 0.7612 0.7949
0.2801 15.0 150 0.7204 {'precision': 0.7103218645948945, 'recall': 0.7911001236093943, 'f1': 0.7485380116959064, 'number': 809} {'precision': 0.3697478991596639, 'recall': 0.3697478991596639, 'f1': 0.3697478991596639, 'number': 119} {'precision': 0.7799126637554585, 'recall': 0.8384976525821596, 'f1': 0.8081447963800905, 'number': 1065} 0.7284 0.7913 0.7585 0.7930

Framework versions

  • Transformers 4.34.0
  • Pytorch 2.0.1+cu118
  • Datasets 2.14.5
  • Tokenizers 0.14.1