metadata
license: mit
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: 0.6953
- Answer: {'precision': 0.7040261153427638, 'recall': 0.799752781211372, 'f1': 0.7488425925925926, 'number': 809}
- Header: {'precision': 0.24516129032258063, 'recall': 0.31932773109243695, 'f1': 0.2773722627737226, 'number': 119}
- Question: {'precision': 0.7894273127753304, 'recall': 0.8413145539906103, 'f1': 0.8145454545454545, 'number': 1065}
- Overall Precision: 0.7157
- Overall Recall: 0.7933
- Overall F1: 0.7525
- Overall Accuracy: 0.8063
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
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Answer | Header | Question | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
|---|---|---|---|---|---|---|---|---|---|---|
| 1.8542 | 1.0 | 10 | 1.6365 | {'precision': 0.01658374792703151, 'recall': 0.012360939431396786, 'f1': 0.014164305949008497, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.21144674085850557, 'recall': 0.12488262910798122, 'f1': 0.15702479338842976, 'number': 1065} | 0.1161 | 0.0718 | 0.0887 | 0.3298 |
| 1.4883 | 2.0 | 20 | 1.3085 | {'precision': 0.17417417417417416, 'recall': 0.21508034610630408, 'f1': 0.19247787610619468, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.37554585152838427, 'recall': 0.48450704225352115, 'f1': 0.4231242312423124, 'number': 1065} | 0.2908 | 0.3462 | 0.3161 | 0.5523 |
| 1.1521 | 3.0 | 30 | 0.9728 | {'precision': 0.452642073778664, 'recall': 0.5611866501854141, 'f1': 0.5011037527593818, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.5948678071539658, 'recall': 0.7183098591549296, 'f1': 0.6507868991918333, 'number': 1065} | 0.53 | 0.6116 | 0.5679 | 0.6953 |
| 0.8808 | 4.0 | 40 | 0.7922 | {'precision': 0.582089552238806, 'recall': 0.723114956736712, 'f1': 0.6449834619625138, 'number': 809} | {'precision': 0.06666666666666667, 'recall': 0.025210084033613446, 'f1': 0.036585365853658534, 'number': 119} | {'precision': 0.6649916247906198, 'recall': 0.7455399061032864, 'f1': 0.7029659141212926, 'number': 1065} | 0.6159 | 0.6934 | 0.6523 | 0.7508 |
| 0.6941 | 5.0 | 50 | 0.7105 | {'precision': 0.6461704422869471, 'recall': 0.7404202719406675, 'f1': 0.6900921658986174, 'number': 809} | {'precision': 0.1875, 'recall': 0.15126050420168066, 'f1': 0.16744186046511625, 'number': 119} | {'precision': 0.651778955336866, 'recall': 0.8084507042253521, 'f1': 0.721709974853311, 'number': 1065} | 0.6305 | 0.7416 | 0.6816 | 0.7809 |
| 0.5844 | 6.0 | 60 | 0.6780 | {'precision': 0.6598360655737705, 'recall': 0.796044499381953, 'f1': 0.7215686274509804, 'number': 809} | {'precision': 0.2079207920792079, 'recall': 0.17647058823529413, 'f1': 0.19090909090909092, 'number': 119} | {'precision': 0.7283633247643531, 'recall': 0.7981220657276995, 'f1': 0.7616487455197132, 'number': 1065} | 0.6751 | 0.7602 | 0.7151 | 0.7926 |
| 0.5021 | 7.0 | 70 | 0.6522 | {'precision': 0.6852248394004282, 'recall': 0.7911001236093943, 'f1': 0.7343660355708548, 'number': 809} | {'precision': 0.20149253731343283, 'recall': 0.226890756302521, 'f1': 0.21343873517786563, 'number': 119} | {'precision': 0.7521514629948365, 'recall': 0.8206572769953052, 'f1': 0.7849124382577458, 'number': 1065} | 0.6910 | 0.7732 | 0.7298 | 0.8036 |
| 0.443 | 8.0 | 80 | 0.6501 | {'precision': 0.6827731092436975, 'recall': 0.8034610630407911, 'f1': 0.7382169222032936, 'number': 809} | {'precision': 0.2542372881355932, 'recall': 0.25210084033613445, 'f1': 0.25316455696202533, 'number': 119} | {'precision': 0.7689003436426117, 'recall': 0.8403755868544601, 'f1': 0.8030506953790938, 'number': 1065} | 0.7050 | 0.7903 | 0.7452 | 0.8060 |
| 0.3917 | 9.0 | 90 | 0.6715 | {'precision': 0.6913319238900634, 'recall': 0.8084054388133498, 'f1': 0.7452991452991453, 'number': 809} | {'precision': 0.25547445255474455, 'recall': 0.29411764705882354, 'f1': 0.2734375, 'number': 119} | {'precision': 0.7811387900355872, 'recall': 0.8244131455399061, 'f1': 0.8021927820922796, 'number': 1065} | 0.7100 | 0.7863 | 0.7462 | 0.8032 |
| 0.3849 | 10.0 | 100 | 0.6725 | {'precision': 0.6908315565031983, 'recall': 0.8009888751545118, 'f1': 0.7418431597023468, 'number': 809} | {'precision': 0.24444444444444444, 'recall': 0.2773109243697479, 'f1': 0.25984251968503935, 'number': 119} | {'precision': 0.7829937998228521, 'recall': 0.8300469483568075, 'f1': 0.805834092980857, 'number': 1065} | 0.7107 | 0.7852 | 0.7461 | 0.8064 |
| 0.3232 | 11.0 | 110 | 0.6747 | {'precision': 0.6918976545842217, 'recall': 0.8022249690976514, 'f1': 0.7429879793932456, 'number': 809} | {'precision': 0.25161290322580643, 'recall': 0.3277310924369748, 'f1': 0.2846715328467153, 'number': 119} | {'precision': 0.7609797297297297, 'recall': 0.8460093896713615, 'f1': 0.8012449977767896, 'number': 1065} | 0.6978 | 0.7973 | 0.7443 | 0.8001 |
| 0.3028 | 12.0 | 120 | 0.6871 | {'precision': 0.700218818380744, 'recall': 0.7911001236093943, 'f1': 0.7428903076030179, 'number': 809} | {'precision': 0.25, 'recall': 0.31092436974789917, 'f1': 0.27715355805243447, 'number': 119} | {'precision': 0.7985611510791367, 'recall': 0.8338028169014085, 'f1': 0.8158015617822691, 'number': 1065} | 0.7199 | 0.7852 | 0.7511 | 0.8042 |
| 0.284 | 13.0 | 130 | 0.6905 | {'precision': 0.697524219590958, 'recall': 0.8009888751545118, 'f1': 0.7456846950517838, 'number': 809} | {'precision': 0.2602739726027397, 'recall': 0.31932773109243695, 'f1': 0.28679245283018867, 'number': 119} | {'precision': 0.7929203539823009, 'recall': 0.8413145539906103, 'f1': 0.8164009111617311, 'number': 1065} | 0.7175 | 0.7938 | 0.7537 | 0.8057 |
| 0.2666 | 14.0 | 140 | 0.6958 | {'precision': 0.6949516648764769, 'recall': 0.799752781211372, 'f1': 0.7436781609195402, 'number': 809} | {'precision': 0.2585034013605442, 'recall': 0.31932773109243695, 'f1': 0.2857142857142857, 'number': 119} | {'precision': 0.7904085257548845, 'recall': 0.8356807511737089, 'f1': 0.8124144226380648, 'number': 1065} | 0.7146 | 0.7903 | 0.7505 | 0.8040 |
| 0.2705 | 15.0 | 150 | 0.6953 | {'precision': 0.7040261153427638, 'recall': 0.799752781211372, 'f1': 0.7488425925925926, 'number': 809} | {'precision': 0.24516129032258063, 'recall': 0.31932773109243695, 'f1': 0.2773722627737226, 'number': 119} | {'precision': 0.7894273127753304, 'recall': 0.8413145539906103, 'f1': 0.8145454545454545, 'number': 1065} | 0.7157 | 0.7933 | 0.7525 | 0.8063 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1