--- 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: 1.3215 - Answer: {'precision': 0.10096818810511757, 'recall': 0.09023485784919653, 'f1': 0.09530026109660573, 'number': 809} - Header: {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} - Question: {'precision': 0.3980815347721823, 'recall': 0.4676056338028169, 'f1': 0.43005181347150256, 'number': 1065} - Overall Precision: 0.2891 - Overall Recall: 0.2865 - Overall F1: 0.2878 - Overall Accuracy: 0.5339 ## 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: 5e-06 - 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 | |:-------------:|:-----:|:----:|:---------------:|:-------------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:| | 1.9471 | 1.0 | 10 | 1.8844 | {'precision': 0.022006141248720572, 'recall': 0.05315203955500618, 'f1': 0.031125588128845458, 'number': 809} | {'precision': 0.00702576112412178, 'recall': 0.05042016806722689, 'f1': 0.012332990750256937, 'number': 119} | {'precision': 0.054583995760466346, 'recall': 0.09671361502347418, 'f1': 0.06978319783197831, 'number': 1065} | 0.0324 | 0.0763 | 0.0455 | 0.2491 | | 1.8584 | 2.0 | 20 | 1.8099 | {'precision': 0.018408941485864562, 'recall': 0.034610630407911, 'f1': 0.024034334763948496, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.08241758241758242, 'recall': 0.11267605633802817, 'f1': 0.09520031733439112, 'number': 1065} | 0.0469 | 0.0743 | 0.0575 | 0.3139 | | 1.7841 | 3.0 | 30 | 1.7444 | {'precision': 0.02190395956192081, 'recall': 0.032138442521631644, 'f1': 0.026052104208416832, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.10752688172043011, 'recall': 0.12206572769953052, 'f1': 0.11433597185576078, 'number': 1065} | 0.0645 | 0.0783 | 0.0707 | 0.3426 | | 1.7255 | 4.0 | 40 | 1.6851 | {'precision': 0.026865671641791045, 'recall': 0.03337453646477132, 'f1': 0.029768467475192944, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.15547024952015356, 'recall': 0.15211267605633802, 'f1': 0.1537731371618415, 'number': 1065} | 0.0922 | 0.0948 | 0.0935 | 0.3647 | | 1.6607 | 5.0 | 50 | 1.6287 | {'precision': 0.036458333333333336, 'recall': 0.04326328800988875, 'f1': 0.03957037874505371, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.2018348623853211, 'recall': 0.20657276995305165, 'f1': 0.20417633410672859, 'number': 1065} | 0.1244 | 0.1279 | 0.1261 | 0.3943 | | 1.6127 | 6.0 | 60 | 1.5738 | {'precision': 0.045, 'recall': 0.05562422744128554, 'f1': 0.04975124378109452, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.24034334763948498, 'recall': 0.26291079812206575, 'f1': 0.25112107623318386, 'number': 1065} | 0.1501 | 0.1631 | 0.1563 | 0.4234 | | 1.5582 | 7.0 | 70 | 1.5242 | {'precision': 0.05465587044534413, 'recall': 0.06674907292954264, 'f1': 0.060100166944908176, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.26282051282051283, 'recall': 0.307981220657277, 'f1': 0.2836143536532642, 'number': 1065} | 0.1708 | 0.1917 | 0.1807 | 0.4483 | | 1.5135 | 8.0 | 80 | 1.4789 | {'precision': 0.05976520811099253, 'recall': 0.069221260815822, 'f1': 0.06414662084765177, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.29073482428115016, 'recall': 0.34178403755868547, 'f1': 0.31419939577039274, 'number': 1065} | 0.1919 | 0.2107 | 0.2009 | 0.4679 | | 1.4676 | 9.0 | 90 | 1.4380 | {'precision': 0.06818181818181818, 'recall': 0.07416563658838071, 'f1': 0.07104795737122557, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.3149480415667466, 'recall': 0.3699530516431925, 'f1': 0.34024179620034545, 'number': 1065} | 0.2130 | 0.2278 | 0.2202 | 0.4851 | | 1.4233 | 10.0 | 100 | 1.4035 | {'precision': 0.07664670658682635, 'recall': 0.07911001236093942, 'f1': 0.0778588807785888, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.3413848631239936, 'recall': 0.39812206572769954, 'f1': 0.3675769397485913, 'number': 1065} | 0.2350 | 0.2449 | 0.2398 | 0.4988 | | 1.3864 | 11.0 | 110 | 1.3744 | {'precision': 0.0810126582278481, 'recall': 0.07911001236093942, 'f1': 0.08005003126954345, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.3583535108958838, 'recall': 0.4169014084507042, 'f1': 0.38541666666666663, 'number': 1065} | 0.2504 | 0.2549 | 0.2526 | 0.5113 | | 1.3746 | 12.0 | 120 | 1.3519 | {'precision': 0.0870712401055409, 'recall': 0.0815822002472188, 'f1': 0.08423739629865987, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.3806818181818182, 'recall': 0.4403755868544601, 'f1': 0.40835872877666524, 'number': 1065} | 0.2688 | 0.2684 | 0.2686 | 0.5175 | | 1.3417 | 13.0 | 130 | 1.3352 | {'precision': 0.09568733153638814, 'recall': 0.08776266996291718, 'f1': 0.09155383623468731, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.39403706688154716, 'recall': 0.4591549295774648, 'f1': 0.4241110147441457, 'number': 1065} | 0.2824 | 0.2810 | 0.2817 | 0.5272 | | 1.3318 | 14.0 | 140 | 1.3254 | {'precision': 0.09686221009549795, 'recall': 0.08776266996291718, 'f1': 0.09208819714656291, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.3942307692307692, 'recall': 0.4619718309859155, 'f1': 0.4254215304798963, 'number': 1065} | 0.2841 | 0.2825 | 0.2833 | 0.5314 | | 1.3086 | 15.0 | 150 | 1.3215 | {'precision': 0.10096818810511757, 'recall': 0.09023485784919653, 'f1': 0.09530026109660573, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.3980815347721823, 'recall': 0.4676056338028169, 'f1': 0.43005181347150256, 'number': 1065} | 0.2891 | 0.2865 | 0.2878 | 0.5339 | ### Framework versions - Transformers 4.31.0 - Pytorch 2.0.1+cu118 - Datasets 2.14.4 - Tokenizers 0.13.3