--- 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: 0.6667 - Answer: {'precision': 0.6578947368421053, 'recall': 0.7725587144622992, 'f1': 0.7106310403638432, 'number': 809} - Header: {'precision': 0.19658119658119658, 'recall': 0.19327731092436976, 'f1': 0.19491525423728814, 'number': 119} - Question: {'precision': 0.7215958369470945, 'recall': 0.7812206572769953, 'f1': 0.7502254283137962, 'number': 1065} - Overall Precision: 0.6667 - Overall Recall: 0.7426 - Overall F1: 0.7026 - Overall Accuracy: 0.7964 ## 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 | |:-------------:|:-----:|:----:|:---------------:|:-----------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:| | 1.7693 | 1.0 | 10 | 1.5725 | {'precision': 0.03488372093023256, 'recall': 0.0407911001236094, 'f1': 0.037606837606837605, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.20363636363636364, 'recall': 0.21032863849765257, 'f1': 0.20692840646651267, 'number': 1065} | 0.1256 | 0.1290 | 0.1273 | 0.3991 | | 1.425 | 2.0 | 20 | 1.2448 | {'precision': 0.12746386333771353, 'recall': 0.11990111248454882, 'f1': 0.1235668789808917, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.44836716681376876, 'recall': 0.47699530516431926, 'f1': 0.46223839854413107, 'number': 1065} | 0.3194 | 0.3036 | 0.3113 | 0.5601 | | 1.1125 | 3.0 | 30 | 0.9760 | {'precision': 0.43318485523385303, 'recall': 0.48084054388133496, 'f1': 0.45577035735207966, 'number': 809} | {'precision': 0.06060606060606061, 'recall': 0.01680672268907563, 'f1': 0.02631578947368421, 'number': 119} | {'precision': 0.6073674752920036, 'recall': 0.6347417840375587, 'f1': 0.620752984389348, 'number': 1065} | 0.5220 | 0.5354 | 0.5286 | 0.6992 | | 0.8731 | 4.0 | 40 | 0.7844 | {'precision': 0.5927835051546392, 'recall': 0.7107540173053152, 'f1': 0.6464305789769533, 'number': 809} | {'precision': 0.12280701754385964, 'recall': 0.058823529411764705, 'f1': 0.07954545454545454, 'number': 119} | {'precision': 0.6381909547738693, 'recall': 0.7154929577464789, 'f1': 0.6746347941567065, 'number': 1065} | 0.6051 | 0.6744 | 0.6379 | 0.7573 | | 0.6964 | 5.0 | 50 | 0.7420 | {'precision': 0.6131868131868132, 'recall': 0.6897404202719407, 'f1': 0.6492146596858639, 'number': 809} | {'precision': 0.17857142857142858, 'recall': 0.12605042016806722, 'f1': 0.14778325123152708, 'number': 119} | {'precision': 0.6419951729686243, 'recall': 0.7492957746478873, 'f1': 0.6915077989601386, 'number': 1065} | 0.6129 | 0.6879 | 0.6482 | 0.7719 | | 0.6156 | 6.0 | 60 | 0.7064 | {'precision': 0.6271008403361344, 'recall': 0.7379480840543882, 'f1': 0.678023850085179, 'number': 809} | {'precision': 0.24, 'recall': 0.15126050420168066, 'f1': 0.18556701030927833, 'number': 119} | {'precision': 0.6932409012131716, 'recall': 0.7511737089201878, 'f1': 0.7210455159981973, 'number': 1065} | 0.6488 | 0.7100 | 0.6780 | 0.7780 | | 0.5557 | 7.0 | 70 | 0.6802 | {'precision': 0.6476793248945147, 'recall': 0.7589616810877626, 'f1': 0.6989186112692088, 'number': 809} | {'precision': 0.22105263157894736, 'recall': 0.17647058823529413, 'f1': 0.19626168224299065, 'number': 119} | {'precision': 0.7050298380221653, 'recall': 0.7765258215962442, 'f1': 0.7390527256479, 'number': 1065} | 0.6597 | 0.7336 | 0.6947 | 0.7915 | | 0.5151 | 8.0 | 80 | 0.6709 | {'precision': 0.6634920634920635, 'recall': 0.7750309023485785, 'f1': 0.7149372862029646, 'number': 809} | {'precision': 0.2072072072072072, 'recall': 0.19327731092436976, 'f1': 0.2, 'number': 119} | {'precision': 0.7220756376429199, 'recall': 0.7708920187793428, 'f1': 0.7456857402361489, 'number': 1065} | 0.6708 | 0.7381 | 0.7028 | 0.7936 | | 0.4746 | 9.0 | 90 | 0.6726 | {'precision': 0.6552462526766595, 'recall': 0.7564894932014833, 'f1': 0.7022375215146299, 'number': 809} | {'precision': 0.21621621621621623, 'recall': 0.20168067226890757, 'f1': 0.20869565217391306, 'number': 119} | {'precision': 0.7148900169204738, 'recall': 0.7934272300469484, 'f1': 0.7521139296840232, 'number': 1065} | 0.6650 | 0.7431 | 0.7019 | 0.7949 | | 0.4849 | 10.0 | 100 | 0.6667 | {'precision': 0.6578947368421053, 'recall': 0.7725587144622992, 'f1': 0.7106310403638432, 'number': 809} | {'precision': 0.19658119658119658, 'recall': 0.19327731092436976, 'f1': 0.19491525423728814, 'number': 119} | {'precision': 0.7215958369470945, 'recall': 0.7812206572769953, 'f1': 0.7502254283137962, 'number': 1065} | 0.6667 | 0.7426 | 0.7026 | 0.7964 | ### Framework versions - Transformers 4.41.2 - Pytorch 2.3.1+cu121 - Datasets 2.19.2 - Tokenizers 0.19.1