--- 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](https://huggingface.co/microsoft/layoutlm-base-uncased) on the funsd dataset. It achieves the following results on the evaluation set: - Loss: 0.7040 - Answer: {'precision': 0.6568109820485745, 'recall': 0.7688504326328801, 'f1': 0.7084282460136675, 'number': 809} - Header: {'precision': 0.2803738317757009, 'recall': 0.25210084033613445, 'f1': 0.2654867256637167, 'number': 119} - Question: {'precision': 0.7009113504556752, 'recall': 0.7943661971830986, 'f1': 0.744718309859155, 'number': 1065} - Overall Precision: 0.6625 - Overall Recall: 0.7516 - Overall F1: 0.7043 - Overall Accuracy: 0.7902 ## 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: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | Answer | Header | Question | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy | |:-------------:|:-----:|:----:|:---------------:|:------------------------------------------------------------------------------------------------------------:|:-----------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:| | 1.8156 | 1.0 | 10 | 1.6000 | {'precision': 0.016967126193001062, 'recall': 0.019777503090234856, 'f1': 0.0182648401826484, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.19947159841479525, 'recall': 0.14178403755868543, 'f1': 0.16575192096597147, 'number': 1065} | 0.0982 | 0.0838 | 0.0904 | 0.3885 | | 1.4929 | 2.0 | 20 | 1.2928 | {'precision': 0.2471213463241807, 'recall': 0.34487021013597036, 'f1': 0.2879256965944273, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.37381275440976935, 'recall': 0.5173708920187794, 'f1': 0.43402914533280823, 'number': 1065} | 0.3189 | 0.4165 | 0.3612 | 0.5918 | | 1.1481 | 3.0 | 30 | 1.0078 | {'precision': 0.37816979051819183, 'recall': 0.42398022249690975, 'f1': 0.3997668997668998, 'number': 809} | {'precision': 0.14705882352941177, 'recall': 0.04201680672268908, 'f1': 0.06535947712418301, 'number': 119} | {'precision': 0.5541346973572038, 'recall': 0.6103286384976526, 'f1': 0.580875781948168, 'number': 1065} | 0.4721 | 0.5008 | 0.4860 | 0.6646 | | 0.9026 | 4.0 | 40 | 0.8847 | {'precision': 0.5041322314049587, 'recall': 0.6786155747836835, 'f1': 0.5785036880927291, 'number': 809} | {'precision': 0.16, 'recall': 0.06722689075630252, 'f1': 0.09467455621301775, 'number': 119} | {'precision': 0.6154513888888888, 'recall': 0.6657276995305165, 'f1': 0.6396030672079386, 'number': 1065} | 0.5526 | 0.6352 | 0.5910 | 0.7209 | | 0.7479 | 5.0 | 50 | 0.7907 | {'precision': 0.6089324618736384, 'recall': 0.6909765142150803, 'f1': 0.6473653734800232, 'number': 809} | {'precision': 0.23076923076923078, 'recall': 0.15126050420168066, 'f1': 0.18274111675126906, 'number': 119} | {'precision': 0.6239600665557404, 'recall': 0.704225352112676, 'f1': 0.6616674018526687, 'number': 1065} | 0.6037 | 0.6658 | 0.6333 | 0.7576 | | 0.651 | 6.0 | 60 | 0.7416 | {'precision': 0.604040404040404, 'recall': 0.7391841779975278, 'f1': 0.6648137854363535, 'number': 809} | {'precision': 0.20238095238095238, 'recall': 0.14285714285714285, 'f1': 0.16748768472906403, 'number': 119} | {'precision': 0.6520376175548589, 'recall': 0.7812206572769953, 'f1': 0.7108073472874841, 'number': 1065} | 0.6157 | 0.7260 | 0.6664 | 0.7732 | | 0.5864 | 7.0 | 70 | 0.7379 | {'precision': 0.6485355648535565, 'recall': 0.7663782447466008, 'f1': 0.7025495750708215, 'number': 809} | {'precision': 0.22772277227722773, 'recall': 0.19327731092436976, 'f1': 0.2090909090909091, 'number': 119} | {'precision': 0.7006861063464837, 'recall': 0.7671361502347418, 'f1': 0.7324069923800985, 'number': 1065} | 0.6568 | 0.7326 | 0.6926 | 0.7746 | | 0.5425 | 8.0 | 80 | 0.7093 | {'precision': 0.6484210526315789, 'recall': 0.761433868974042, 'f1': 0.7003979533826037, 'number': 809} | {'precision': 0.25925925925925924, 'recall': 0.23529411764705882, 'f1': 0.24669603524229072, 'number': 119} | {'precision': 0.6843800322061192, 'recall': 0.7981220657276995, 'f1': 0.7368877329865627, 'number': 1065} | 0.6496 | 0.7496 | 0.6960 | 0.7901 | | 0.4986 | 9.0 | 90 | 0.7080 | {'precision': 0.6553911205073996, 'recall': 0.7663782447466008, 'f1': 0.7065527065527065, 'number': 809} | {'precision': 0.2857142857142857, 'recall': 0.25210084033613445, 'f1': 0.26785714285714285, 'number': 119} | {'precision': 0.7062761506276151, 'recall': 0.7924882629107981, 'f1': 0.7469026548672565, 'number': 1065} | 0.6652 | 0.7496 | 0.7049 | 0.7881 | | 0.481 | 10.0 | 100 | 0.7040 | {'precision': 0.6568109820485745, 'recall': 0.7688504326328801, 'f1': 0.7084282460136675, 'number': 809} | {'precision': 0.2803738317757009, 'recall': 0.25210084033613445, 'f1': 0.2654867256637167, 'number': 119} | {'precision': 0.7009113504556752, 'recall': 0.7943661971830986, 'f1': 0.744718309859155, 'number': 1065} | 0.6625 | 0.7516 | 0.7043 | 0.7902 | ### Framework versions - Transformers 4.31.0 - Pytorch 2.0.1+cu118 - Datasets 2.14.2 - Tokenizers 0.13.3