--- 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: 0.7271 - Answer: {'precision': 0.7216157205240175, 'recall': 0.8170580964153276, 'f1': 0.766376811594203, 'number': 809} - Header: {'precision': 0.3046875, 'recall': 0.3277310924369748, 'f1': 0.31578947368421056, 'number': 119} - Question: {'precision': 0.7844905320108205, 'recall': 0.8169014084507042, 'f1': 0.8003679852805888, 'number': 1065} - Overall Precision: 0.7292 - Overall Recall: 0.7878 - Overall F1: 0.7574 - Overall Accuracy: 0.8010 ## 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: 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 | |:-------------:|:-----:|:----:|:---------------:|:-----------------------------------------------------------------------------------------------------------:|:-----------------------------------------------------------------------------------------------------------:|:-----------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:| | 1.7704 | 1.0 | 10 | 1.5853 | {'precision': 0.0226628895184136, 'recall': 0.019777503090234856, 'f1': 0.02112211221122112, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.26928471248246844, 'recall': 0.18028169014084508, 'f1': 0.2159730033745782, 'number': 1065} | 0.1466 | 0.1044 | 0.1219 | 0.3606 | | 1.4598 | 2.0 | 20 | 1.2597 | {'precision': 0.14809384164222875, 'recall': 0.12484548825710753, 'f1': 0.13547954393024816, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.3946969696969697, 'recall': 0.4892018779342723, 'f1': 0.4368972746331237, 'number': 1065} | 0.3107 | 0.3121 | 0.3114 | 0.5801 | | 1.1343 | 3.0 | 30 | 0.9794 | {'precision': 0.45892018779342725, 'recall': 0.48331273176761436, 'f1': 0.47080072245635163, 'number': 809} | {'precision': 0.09523809523809523, 'recall': 0.01680672268907563, 'f1': 0.02857142857142857, 'number': 119} | {'precision': 0.578684429641965, 'recall': 0.6525821596244131, 'f1': 0.6134157105030891, 'number': 1065} | 0.5246 | 0.5459 | 0.5350 | 0.7059 | | 0.8715 | 4.0 | 40 | 0.8169 | {'precision': 0.5791245791245792, 'recall': 0.6378244746600742, 'f1': 0.6070588235294119, 'number': 809} | {'precision': 0.20833333333333334, 'recall': 0.08403361344537816, 'f1': 0.11976047904191618, 'number': 119} | {'precision': 0.6848816029143898, 'recall': 0.7061032863849765, 'f1': 0.6953305594082293, 'number': 1065} | 0.6274 | 0.6412 | 0.6342 | 0.7404 | | 0.6755 | 5.0 | 50 | 0.7034 | {'precision': 0.6616541353383458, 'recall': 0.761433868974042, 'f1': 0.7080459770114943, 'number': 809} | {'precision': 0.2, 'recall': 0.14285714285714285, 'f1': 0.16666666666666666, 'number': 119} | {'precision': 0.6862745098039216, 'recall': 0.7887323943661971, 'f1': 0.7339449541284403, 'number': 1065} | 0.6576 | 0.7391 | 0.6960 | 0.7888 | | 0.563 | 6.0 | 60 | 0.6924 | {'precision': 0.6618998978549541, 'recall': 0.8009888751545118, 'f1': 0.7248322147651008, 'number': 809} | {'precision': 0.2191780821917808, 'recall': 0.13445378151260504, 'f1': 0.16666666666666669, 'number': 119} | {'precision': 0.7177489177489178, 'recall': 0.7784037558685446, 'f1': 0.7468468468468469, 'number': 1065} | 0.6765 | 0.7491 | 0.7110 | 0.7869 | | 0.4764 | 7.0 | 70 | 0.6676 | {'precision': 0.7162011173184357, 'recall': 0.792336217552534, 'f1': 0.7523474178403756, 'number': 809} | {'precision': 0.26126126126126126, 'recall': 0.24369747899159663, 'f1': 0.25217391304347825, 'number': 119} | {'precision': 0.7538726333907056, 'recall': 0.8225352112676056, 'f1': 0.7867085765603951, 'number': 1065} | 0.7131 | 0.7757 | 0.7431 | 0.8032 | | 0.4205 | 8.0 | 80 | 0.6759 | {'precision': 0.7108953613807982, 'recall': 0.8145859085290482, 'f1': 0.7592165898617511, 'number': 809} | {'precision': 0.2564102564102564, 'recall': 0.25210084033613445, 'f1': 0.2542372881355932, 'number': 119} | {'precision': 0.7594501718213058, 'recall': 0.8300469483568075, 'f1': 0.7931807985643785, 'number': 1065} | 0.7124 | 0.7893 | 0.7489 | 0.8005 | | 0.3675 | 9.0 | 90 | 0.6917 | {'precision': 0.7132034632034632, 'recall': 0.8145859085290482, 'f1': 0.7605308713214081, 'number': 809} | {'precision': 0.25984251968503935, 'recall': 0.2773109243697479, 'f1': 0.2682926829268293, 'number': 119} | {'precision': 0.7740213523131673, 'recall': 0.8169014084507042, 'f1': 0.7948835084513477, 'number': 1065} | 0.7182 | 0.7837 | 0.7495 | 0.7982 | | 0.3596 | 10.0 | 100 | 0.6906 | {'precision': 0.7193932827735645, 'recall': 0.8207663782447466, 'f1': 0.766743648960739, 'number': 809} | {'precision': 0.3, 'recall': 0.2773109243697479, 'f1': 0.28820960698689957, 'number': 119} | {'precision': 0.7866786678667866, 'recall': 0.8206572769953052, 'f1': 0.8033088235294117, 'number': 1065} | 0.7327 | 0.7883 | 0.7595 | 0.8061 | | 0.3121 | 11.0 | 110 | 0.6999 | {'precision': 0.7300884955752213, 'recall': 0.8158220024721878, 'f1': 0.7705779334500875, 'number': 809} | {'precision': 0.3082706766917293, 'recall': 0.3445378151260504, 'f1': 0.3253968253968254, 'number': 119} | {'precision': 0.7694974003466204, 'recall': 0.8338028169014085, 'f1': 0.8003605227579991, 'number': 1065} | 0.7252 | 0.7973 | 0.7596 | 0.8035 | | 0.2902 | 12.0 | 120 | 0.7153 | {'precision': 0.7124183006535948, 'recall': 0.8084054388133498, 'f1': 0.7573827446438912, 'number': 809} | {'precision': 0.3185840707964602, 'recall': 0.3025210084033613, 'f1': 0.3103448275862069, 'number': 119} | {'precision': 0.7887067395264117, 'recall': 0.8131455399061033, 'f1': 0.8007397133610726, 'number': 1065} | 0.7309 | 0.7807 | 0.7550 | 0.8029 | | 0.2776 | 13.0 | 130 | 0.7184 | {'precision': 0.728587319243604, 'recall': 0.8096415327564895, 'f1': 0.7669789227166277, 'number': 809} | {'precision': 0.3, 'recall': 0.3277310924369748, 'f1': 0.3132530120481928, 'number': 119} | {'precision': 0.7759226713532513, 'recall': 0.8291079812206573, 'f1': 0.8016341352700863, 'number': 1065} | 0.7277 | 0.7913 | 0.7582 | 0.8015 | | 0.2604 | 14.0 | 140 | 0.7218 | {'precision': 0.7272727272727273, 'recall': 0.8108776266996292, 'f1': 0.766803039158387, 'number': 809} | {'precision': 0.31746031746031744, 'recall': 0.33613445378151263, 'f1': 0.32653061224489793, 'number': 119} | {'precision': 0.7809439002671416, 'recall': 0.8234741784037559, 'f1': 0.8016453382084096, 'number': 1065} | 0.7313 | 0.7893 | 0.7592 | 0.8021 | | 0.2644 | 15.0 | 150 | 0.7271 | {'precision': 0.7216157205240175, 'recall': 0.8170580964153276, 'f1': 0.766376811594203, 'number': 809} | {'precision': 0.3046875, 'recall': 0.3277310924369748, 'f1': 0.31578947368421056, 'number': 119} | {'precision': 0.7844905320108205, 'recall': 0.8169014084507042, 'f1': 0.8003679852805888, 'number': 1065} | 0.7292 | 0.7878 | 0.7574 | 0.8010 | ### Framework versions - Transformers 4.41.2 - Pytorch 2.3.0+cu121 - Datasets 2.19.2 - Tokenizers 0.19.1