--- license: mit base_model: microsoft/layoutlm-base-uncased tags: - generated_from_trainer datasets: - funsd model-index: - name: layoutlm-funsd1 results: [] --- # layoutlm-funsd1 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: 0.6576 - Answer: {'precision': 0.6760869565217391, 'recall': 0.7688504326328801, 'f1': 0.719491035280509, 'number': 809} - Header: {'precision': 0.29473684210526313, 'recall': 0.23529411764705882, 'f1': 0.2616822429906542, 'number': 119} - Question: {'precision': 0.7385398981324278, 'recall': 0.8169014084507042, 'f1': 0.7757467677218011, 'number': 1065} - Overall Precision: 0.6931 - Overall Recall: 0.7627 - Overall F1: 0.7262 - Overall Accuracy: 0.7966 ## 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 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Answer | Header | Question | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy | |:-------------:|:-----:|:----:|:---------------:|:--------------------------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:| | 1.8473 | 1.0 | 10 | 1.5928 | {'precision': 0.018163471241170535, 'recall': 0.022249690976514216, 'f1': 0.020000000000000004, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.22706209453197404, 'recall': 0.2300469483568075, 'f1': 0.228544776119403, 'number': 1065} | 0.1271 | 0.1320 | 0.1295 | 0.3941 | | 1.4704 | 2.0 | 20 | 1.2787 | {'precision': 0.11602870813397129, 'recall': 0.11990111248454882, 'f1': 0.11793313069908813, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.4026946107784431, 'recall': 0.5051643192488263, 'f1': 0.4481466055810079, 'number': 1065} | 0.2924 | 0.3186 | 0.3049 | 0.5625 | | 1.1341 | 3.0 | 30 | 1.0026 | {'precision': 0.3333333333333333, 'recall': 0.33127317676143386, 'f1': 0.33230006199628026, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.5989804587935429, 'recall': 0.6619718309859155, 'f1': 0.6289027653880465, 'number': 1065} | 0.4831 | 0.4882 | 0.4857 | 0.6604 | | 0.8967 | 4.0 | 40 | 0.8387 | {'precision': 0.571563981042654, 'recall': 0.7453646477132262, 'f1': 0.6469957081545066, 'number': 809} | {'precision': 0.06976744186046512, 'recall': 0.025210084033613446, 'f1': 0.037037037037037035, 'number': 119} | {'precision': 0.6548748921484038, 'recall': 0.7126760563380282, 'f1': 0.6825539568345323, 'number': 1065} | 0.6048 | 0.6849 | 0.6424 | 0.7382 | | 0.723 | 5.0 | 50 | 0.7520 | {'precision': 0.5984174085064293, 'recall': 0.7478368355995055, 'f1': 0.6648351648351648, 'number': 809} | {'precision': 0.1935483870967742, 'recall': 0.10084033613445378, 'f1': 0.13259668508287292, 'number': 119} | {'precision': 0.6901041666666666, 'recall': 0.7464788732394366, 'f1': 0.7171853856562922, 'number': 1065} | 0.6346 | 0.7085 | 0.6695 | 0.7621 | | 0.6196 | 6.0 | 60 | 0.7171 | {'precision': 0.6231003039513677, 'recall': 0.7601977750309024, 'f1': 0.6848552338530067, 'number': 809} | {'precision': 0.2125, 'recall': 0.14285714285714285, 'f1': 0.1708542713567839, 'number': 119} | {'precision': 0.7221238938053097, 'recall': 0.7661971830985915, 'f1': 0.743507972665148, 'number': 1065} | 0.6591 | 0.7265 | 0.6912 | 0.7734 | | 0.5747 | 7.0 | 70 | 0.6993 | {'precision': 0.6506410256410257, 'recall': 0.7527812113720643, 'f1': 0.6979942693409743, 'number': 809} | {'precision': 0.2558139534883721, 'recall': 0.18487394957983194, 'f1': 0.21463414634146344, 'number': 119} | {'precision': 0.6894060995184591, 'recall': 0.8065727699530516, 'f1': 0.7434011250540891, 'number': 1065} | 0.6570 | 0.7476 | 0.6994 | 0.7841 | | 0.5292 | 8.0 | 80 | 0.6785 | {'precision': 0.6484536082474227, 'recall': 0.7775030902348579, 'f1': 0.7071388420460932, 'number': 809} | {'precision': 0.29069767441860467, 'recall': 0.21008403361344538, 'f1': 0.24390243902439027, 'number': 119} | {'precision': 0.7459893048128342, 'recall': 0.7859154929577464, 'f1': 0.7654320987654322, 'number': 1065} | 0.6846 | 0.7481 | 0.7149 | 0.7893 | | 0.4862 | 9.0 | 90 | 0.6637 | {'precision': 0.658008658008658, 'recall': 0.7515451174289246, 'f1': 0.7016733987305251, 'number': 809} | {'precision': 0.28125, 'recall': 0.226890756302521, 'f1': 0.2511627906976744, 'number': 119} | {'precision': 0.7287853577371048, 'recall': 0.8225352112676056, 'f1': 0.7728275253639171, 'number': 1065} | 0.6800 | 0.7582 | 0.7170 | 0.7931 | | 0.4795 | 10.0 | 100 | 0.6576 | {'precision': 0.6760869565217391, 'recall': 0.7688504326328801, 'f1': 0.719491035280509, 'number': 809} | {'precision': 0.29473684210526313, 'recall': 0.23529411764705882, 'f1': 0.2616822429906542, 'number': 119} | {'precision': 0.7385398981324278, 'recall': 0.8169014084507042, 'f1': 0.7757467677218011, 'number': 1065} | 0.6931 | 0.7627 | 0.7262 | 0.7966 | ### Framework versions - Transformers 4.41.2 - Pytorch 2.3.1+cu121 - Datasets 2.19.2 - Tokenizers 0.19.1