--- library_name: transformers 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](https://huggingface.co/microsoft/layoutlm-base-uncased) on the funsd dataset. It achieves the following results on the evaluation set: - Loss: 0.6795 - Answer: {'precision': 0.7247807017543859, 'recall': 0.8170580964153276, 'f1': 0.768158047646717, 'number': 809} - Header: {'precision': 0.3208955223880597, 'recall': 0.36134453781512604, 'f1': 0.33992094861660077, 'number': 119} - Question: {'precision': 0.7793721973094171, 'recall': 0.815962441314554, 'f1': 0.7972477064220184, 'number': 1065} - Overall Precision: 0.7279 - Overall Recall: 0.7893 - Overall F1: 0.7573 - Overall Accuracy: 0.8097 ## 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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - 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 | |:-------------:|:-----:|:----:|:---------------:|:----------------------------------------------------------------------------------------------------------:|:-----------------------------------------------------------------------------------------------------------:|:----------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:| | 1.7157 | 1.0 | 10 | 1.4996 | {'precision': 0.07431693989071038, 'recall': 0.08405438813349815, 'f1': 0.0788863109048724, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.22849213691026826, 'recall': 0.231924882629108, 'f1': 0.23019571295433364, 'number': 1065} | 0.1578 | 0.1581 | 0.1579 | 0.4464 | | 1.3388 | 2.0 | 20 | 1.1680 | {'precision': 0.2980891719745223, 'recall': 0.2892459826946848, 'f1': 0.2936010037641154, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.49192245557350567, 'recall': 0.571830985915493, 'f1': 0.5288753799392097, 'number': 1065} | 0.4167 | 0.4230 | 0.4198 | 0.6170 | | 1.0157 | 3.0 | 30 | 0.8917 | {'precision': 0.5579302587176603, 'recall': 0.6131025957972805, 'f1': 0.5842167255594817, 'number': 809} | {'precision': 0.13513513513513514, 'recall': 0.04201680672268908, 'f1': 0.06410256410256411, 'number': 119} | {'precision': 0.5931528662420382, 'recall': 0.6995305164319249, 'f1': 0.6419646704006894, 'number': 1065} | 0.5710 | 0.6252 | 0.5969 | 0.7325 | | 0.7726 | 4.0 | 40 | 0.7448 | {'precision': 0.6359875904860393, 'recall': 0.7601977750309024, 'f1': 0.6925675675675675, 'number': 809} | {'precision': 0.29850746268656714, 'recall': 0.16806722689075632, 'f1': 0.21505376344086022, 'number': 119} | {'precision': 0.691696113074205, 'recall': 0.7352112676056338, 'f1': 0.7127901684114702, 'number': 1065} | 0.6547 | 0.7115 | 0.6819 | 0.7725 | | 0.6355 | 5.0 | 50 | 0.6844 | {'precision': 0.6714129244249726, 'recall': 0.757725587144623, 'f1': 0.7119628339140535, 'number': 809} | {'precision': 0.3170731707317073, 'recall': 0.2184873949579832, 'f1': 0.25870646766169153, 'number': 119} | {'precision': 0.705, 'recall': 0.7943661971830986, 'f1': 0.7470198675496689, 'number': 1065} | 0.6765 | 0.7451 | 0.7092 | 0.7944 | | 0.5353 | 6.0 | 60 | 0.6699 | {'precision': 0.6676860346585117, 'recall': 0.8096415327564895, 'f1': 0.7318435754189945, 'number': 809} | {'precision': 0.3068181818181818, 'recall': 0.226890756302521, 'f1': 0.2608695652173913, 'number': 119} | {'precision': 0.7171717171717171, 'recall': 0.8, 'f1': 0.7563249001331558, 'number': 1065} | 0.6797 | 0.7697 | 0.7219 | 0.7951 | | 0.4614 | 7.0 | 70 | 0.6517 | {'precision': 0.7006507592190889, 'recall': 0.7985166872682324, 'f1': 0.7463893703061815, 'number': 809} | {'precision': 0.26495726495726496, 'recall': 0.2605042016806723, 'f1': 0.2627118644067797, 'number': 119} | {'precision': 0.7355442176870748, 'recall': 0.812206572769953, 'f1': 0.7719767960731816, 'number': 1065} | 0.6962 | 0.7737 | 0.7329 | 0.8045 | | 0.4076 | 8.0 | 80 | 0.6567 | {'precision': 0.7194719471947195, 'recall': 0.8084054388133498, 'f1': 0.761350407450524, 'number': 809} | {'precision': 0.2713178294573643, 'recall': 0.29411764705882354, 'f1': 0.28225806451612906, 'number': 119} | {'precision': 0.7508561643835616, 'recall': 0.8234741784037559, 'f1': 0.7854903716972683, 'number': 1065} | 0.7099 | 0.7858 | 0.7459 | 0.8056 | | 0.3664 | 9.0 | 90 | 0.6529 | {'precision': 0.7176724137931034, 'recall': 0.823238566131026, 'f1': 0.7668393782383419, 'number': 809} | {'precision': 0.265625, 'recall': 0.2857142857142857, 'f1': 0.27530364372469635, 'number': 119} | {'precision': 0.7686768676867687, 'recall': 0.8018779342723005, 'f1': 0.7849264705882354, 'number': 1065} | 0.7171 | 0.7797 | 0.7471 | 0.8098 | | 0.3515 | 10.0 | 100 | 0.6537 | {'precision': 0.7129032258064516, 'recall': 0.8195302843016069, 'f1': 0.7625071880391028, 'number': 809} | {'precision': 0.32075471698113206, 'recall': 0.2857142857142857, 'f1': 0.30222222222222217, 'number': 119} | {'precision': 0.7788632326820604, 'recall': 0.8234741784037559, 'f1': 0.8005476951163851, 'number': 1065} | 0.7280 | 0.7898 | 0.7576 | 0.8157 | | 0.3077 | 11.0 | 110 | 0.6694 | {'precision': 0.7241379310344828, 'recall': 0.8046971569839307, 'f1': 0.7622950819672132, 'number': 809} | {'precision': 0.302158273381295, 'recall': 0.35294117647058826, 'f1': 0.3255813953488373, 'number': 119} | {'precision': 0.7599653379549394, 'recall': 0.8234741784037559, 'f1': 0.790446146913024, 'number': 1065} | 0.7162 | 0.7878 | 0.7503 | 0.8078 | | 0.2941 | 12.0 | 120 | 0.6687 | {'precision': 0.7135076252723311, 'recall': 0.8096415327564895, 'f1': 0.7585408222350897, 'number': 809} | {'precision': 0.3053435114503817, 'recall': 0.33613445378151263, 'f1': 0.32000000000000006, 'number': 119} | {'precision': 0.7793721973094171, 'recall': 0.815962441314554, 'f1': 0.7972477064220184, 'number': 1065} | 0.7227 | 0.7847 | 0.7525 | 0.8118 | | 0.2726 | 13.0 | 130 | 0.6769 | {'precision': 0.720348204570185, 'recall': 0.8182941903584673, 'f1': 0.7662037037037037, 'number': 809} | {'precision': 0.33064516129032256, 'recall': 0.3445378151260504, 'f1': 0.33744855967078186, 'number': 119} | {'precision': 0.7713280562884784, 'recall': 0.8234741784037559, 'f1': 0.7965485921889193, 'number': 1065} | 0.7248 | 0.7928 | 0.7572 | 0.8095 | | 0.2575 | 14.0 | 140 | 0.6821 | {'precision': 0.7238723872387238, 'recall': 0.8133498145859085, 'f1': 0.7660069848661233, 'number': 809} | {'precision': 0.3333333333333333, 'recall': 0.3697478991596639, 'f1': 0.350597609561753, 'number': 119} | {'precision': 0.7810545129579982, 'recall': 0.8206572769953052, 'f1': 0.8003663003663004, 'number': 1065} | 0.7296 | 0.7908 | 0.7590 | 0.8095 | | 0.2563 | 15.0 | 150 | 0.6795 | {'precision': 0.7247807017543859, 'recall': 0.8170580964153276, 'f1': 0.768158047646717, 'number': 809} | {'precision': 0.3208955223880597, 'recall': 0.36134453781512604, 'f1': 0.33992094861660077, 'number': 119} | {'precision': 0.7793721973094171, 'recall': 0.815962441314554, 'f1': 0.7972477064220184, 'number': 1065} | 0.7279 | 0.7893 | 0.7573 | 0.8097 | ### Framework versions - Transformers 4.47.0 - Pytorch 2.5.1+cpu - Datasets 3.2.0 - Tokenizers 0.21.0