--- 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: 1.3293 - Answer: {'precision': 0.11451135241855874, 'recall': 0.1433868974042027, 'f1': 0.12733260153677278, 'number': 809} - Header: {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} - Question: {'precision': 0.41704374057315236, 'recall': 0.5192488262910798, 'f1': 0.46256796319531585, 'number': 1065} - Overall Precision: 0.2860 - Overall Recall: 0.3357 - Overall F1: 0.3089 - Overall Accuracy: 0.5623 ## 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 | |:-------------:|:-----:|:----:|:---------------:|:-------------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:| | 1.9774 | 1.0 | 10 | 1.9285 | {'precision': 0.018331226295828066, 'recall': 0.03584672435105068, 'f1': 0.024257632789627767, 'number': 809} | {'precision': 0.00787878787878788, 'recall': 0.1092436974789916, 'f1': 0.014697569248162805, 'number': 119} | {'precision': 0.06559356136820925, 'recall': 0.15305164319248826, 'f1': 0.09183098591549295, 'number': 1065} | 0.0359 | 0.1029 | 0.0532 | 0.1843 | | 1.8918 | 2.0 | 20 | 1.8488 | {'precision': 0.02769385699899295, 'recall': 0.06798516687268233, 'f1': 0.03935599284436494, 'number': 809} | {'precision': 0.003703703703703704, 'recall': 0.008403361344537815, 'f1': 0.0051413881748071984, 'number': 119} | {'precision': 0.07554585152838428, 'recall': 0.1624413145539906, 'f1': 0.10312965722801788, 'number': 1065} | 0.0504 | 0.1149 | 0.0700 | 0.2606 | | 1.8117 | 3.0 | 30 | 1.7797 | {'precision': 0.02564102564102564, 'recall': 0.0580964153275649, 'f1': 0.03557910673732021, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.0943496801705757, 'recall': 0.16619718309859155, 'f1': 0.120367222033322, 'number': 1065} | 0.0601 | 0.1124 | 0.0783 | 0.3026 | | 1.7441 | 4.0 | 40 | 1.7198 | {'precision': 0.019028871391076115, 'recall': 0.03584672435105068, 'f1': 0.024860694384912133, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.12127512127512127, 'recall': 0.1643192488262911, 'f1': 0.13955342902711323, 'number': 1065} | 0.0686 | 0.1024 | 0.0822 | 0.3324 | | 1.6818 | 5.0 | 50 | 1.6641 | {'precision': 0.0196078431372549, 'recall': 0.03337453646477132, 'f1': 0.024702653247941447, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.15128593040847202, 'recall': 0.18779342723004694, 'f1': 0.16757436112274823, 'number': 1065} | 0.0841 | 0.1139 | 0.0968 | 0.3537 | | 1.6335 | 6.0 | 60 | 1.6097 | {'precision': 0.02643171806167401, 'recall': 0.04449938195302843, 'f1': 0.03316444035006909, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.18782870022539444, 'recall': 0.2347417840375587, 'f1': 0.20868113522537562, 'number': 1065} | 0.1062 | 0.1435 | 0.1221 | 0.3821 | | 1.5742 | 7.0 | 70 | 1.5578 | {'precision': 0.033409263477600606, 'recall': 0.054388133498145856, 'f1': 0.04139228598306679, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.22088068181818182, 'recall': 0.292018779342723, 'f1': 0.2515163768701982, 'number': 1065} | 0.1303 | 0.1781 | 0.1505 | 0.4189 | | 1.5302 | 8.0 | 80 | 1.5083 | {'precision': 0.0456656346749226, 'recall': 0.07292954264524104, 'f1': 0.05616373155640171, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.24610169491525424, 'recall': 0.3408450704225352, 'f1': 0.2858267716535433, 'number': 1065} | 0.1525 | 0.2117 | 0.1773 | 0.4559 | | 1.4774 | 9.0 | 90 | 1.4639 | {'precision': 0.05325914149443561, 'recall': 0.08281829419035847, 'f1': 0.0648282535074988, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.28843537414965986, 'recall': 0.39812206572769954, 'f1': 0.33451676528599605, 'number': 1065} | 0.1800 | 0.2464 | 0.2080 | 0.4889 | | 1.4389 | 10.0 | 100 | 1.4263 | {'precision': 0.059574468085106386, 'recall': 0.0865265760197775, 'f1': 0.07056451612903225, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.32748948106591863, 'recall': 0.4384976525821596, 'f1': 0.3749498193496587, 'number': 1065} | 0.2065 | 0.2694 | 0.2338 | 0.5120 | | 1.4007 | 11.0 | 110 | 1.3933 | {'precision': 0.07123534715960325, 'recall': 0.09765142150803462, 'f1': 0.08237747653806049, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.360773085182534, 'recall': 0.4732394366197183, 'f1': 0.40942323314378554, 'number': 1065} | 0.2326 | 0.2925 | 0.2592 | 0.5334 | | 1.3866 | 12.0 | 120 | 1.3665 | {'precision': 0.09439252336448598, 'recall': 0.12484548825710753, 'f1': 0.10750399148483236, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.38648052902277735, 'recall': 0.49389671361502346, 'f1': 0.4336356141797197, 'number': 1065} | 0.2579 | 0.3146 | 0.2835 | 0.5428 | | 1.3482 | 13.0 | 130 | 1.3469 | {'precision': 0.10622009569377991, 'recall': 0.13720642768850433, 'f1': 0.11974110032362459, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.40044411547002223, 'recall': 0.507981220657277, 'f1': 0.44784768211920534, 'number': 1065} | 0.2721 | 0.3271 | 0.2971 | 0.5537 | | 1.3355 | 14.0 | 140 | 1.3345 | {'precision': 0.11078431372549019, 'recall': 0.13967861557478367, 'f1': 0.12356478950246036, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.4114114114114114, 'recall': 0.5145539906103287, 'f1': 0.4572382144347101, 'number': 1065} | 0.2810 | 0.3317 | 0.3043 | 0.5588 | | 1.3066 | 15.0 | 150 | 1.3293 | {'precision': 0.11451135241855874, 'recall': 0.1433868974042027, 'f1': 0.12733260153677278, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.41704374057315236, 'recall': 0.5192488262910798, 'f1': 0.46256796319531585, 'number': 1065} | 0.2860 | 0.3357 | 0.3089 | 0.5623 | ### Framework versions - Transformers 4.31.0 - Pytorch 2.0.1+cu118 - Datasets 2.14.4 - Tokenizers 0.13.3