metadata
license: mit
base_model: microsoft/layoutlm-base-uncased
tags:
- generated_from_trainer
model-index:
- name: layoutlm-funsd
results: []
layoutlm-funsd
This model is a fine-tuned version of microsoft/layoutlm-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6624
- Answer: {'precision': 0.7003222341568206, 'recall': 0.8059332509270705, 'f1': 0.7494252873563217, 'number': 809}
- Header: {'precision': 0.3148148148148148, 'recall': 0.2857142857142857, 'f1': 0.29955947136563876, 'number': 119}
- Question: {'precision': 0.7602441150828247, 'recall': 0.8187793427230047, 'f1': 0.7884267631103073, 'number': 1065}
- Overall Precision: 0.7127
- Overall Recall: 0.7817
- Overall F1: 0.7456
- Overall Accuracy: 0.8098
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 |
|---|---|---|---|---|---|---|---|---|---|---|
| 1.8207 | 1.0 | 10 | 1.6331 | {'precision': 0.01676829268292683, 'recall': 0.013597033374536464, 'f1': 0.015017064846416382, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.21189024390243902, 'recall': 0.13051643192488263, 'f1': 0.16153399186519465, 'number': 1065} | 0.1143 | 0.0753 | 0.0908 | 0.3429 |
| 1.4867 | 2.0 | 20 | 1.3144 | {'precision': 0.13937282229965156, 'recall': 0.14833127317676142, 'f1': 0.14371257485029942, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.4092178770949721, 'recall': 0.5502347417840375, 'f1': 0.4693632358830597, 'number': 1065} | 0.3079 | 0.3542 | 0.3294 | 0.5753 |
| 1.1706 | 3.0 | 30 | 1.0082 | {'precision': 0.4507042253521127, 'recall': 0.553770086526576, 'f1': 0.4969495285635052, 'number': 809} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119} | {'precision': 0.5885810243492863, 'recall': 0.6582159624413145, 'f1': 0.6214539007092199, 'number': 1065} | 0.5237 | 0.5765 | 0.5488 | 0.6721 |
| 0.8874 | 4.0 | 40 | 0.8115 | {'precision': 0.6029106029106029, 'recall': 0.7169344870210136, 'f1': 0.6549971767363072, 'number': 809} | {'precision': 0.05714285714285714, 'recall': 0.01680672268907563, 'f1': 0.025974025974025972, 'number': 119} | {'precision': 0.649792531120332, 'recall': 0.7352112676056338, 'f1': 0.6898678414096917, 'number': 1065} | 0.6199 | 0.6849 | 0.6508 | 0.7517 |
| 0.7072 | 5.0 | 50 | 0.7206 | {'precision': 0.6341948310139165, 'recall': 0.788627935723115, 'f1': 0.7030303030303031, 'number': 809} | {'precision': 0.18032786885245902, 'recall': 0.09243697478991597, 'f1': 0.12222222222222223, 'number': 119} | {'precision': 0.696551724137931, 'recall': 0.7586854460093897, 'f1': 0.7262921348314607, 'number': 1065} | 0.6542 | 0.7311 | 0.6905 | 0.7725 |
| 0.5896 | 6.0 | 60 | 0.6813 | {'precision': 0.6571428571428571, 'recall': 0.796044499381953, 'f1': 0.7199552822806037, 'number': 809} | {'precision': 0.1746031746031746, 'recall': 0.09243697478991597, 'f1': 0.12087912087912087, 'number': 119} | {'precision': 0.7217981340118744, 'recall': 0.7990610328638498, 'f1': 0.7584670231729055, 'number': 1065} | 0.6778 | 0.7556 | 0.7146 | 0.7867 |
| 0.5193 | 7.0 | 70 | 0.6605 | {'precision': 0.6949516648764769, 'recall': 0.799752781211372, 'f1': 0.7436781609195402, 'number': 809} | {'precision': 0.20618556701030927, 'recall': 0.16806722689075632, 'f1': 0.1851851851851852, 'number': 119} | {'precision': 0.734468085106383, 'recall': 0.8103286384976526, 'f1': 0.7705357142857142, 'number': 1065} | 0.6945 | 0.7677 | 0.7293 | 0.7979 |
| 0.4591 | 8.0 | 80 | 0.6473 | {'precision': 0.6922246220302376, 'recall': 0.792336217552534, 'f1': 0.7389048991354467, 'number': 809} | {'precision': 0.24, 'recall': 0.20168067226890757, 'f1': 0.2191780821917808, 'number': 119} | {'precision': 0.7382154882154882, 'recall': 0.8234741784037559, 'f1': 0.7785175321793164, 'number': 1065} | 0.6965 | 0.7737 | 0.7331 | 0.8059 |
| 0.3939 | 9.0 | 90 | 0.6369 | {'precision': 0.6886291179596175, 'recall': 0.8009888751545118, 'f1': 0.7405714285714285, 'number': 809} | {'precision': 0.2777777777777778, 'recall': 0.25210084033613445, 'f1': 0.2643171806167401, 'number': 119} | {'precision': 0.7515047291487532, 'recall': 0.8206572769953052, 'f1': 0.784560143626571, 'number': 1065} | 0.7016 | 0.7787 | 0.7382 | 0.8088 |
| 0.3604 | 10.0 | 100 | 0.6514 | {'precision': 0.6954643628509719, 'recall': 0.796044499381953, 'f1': 0.7423631123919308, 'number': 809} | {'precision': 0.29, 'recall': 0.24369747899159663, 'f1': 0.2648401826484018, 'number': 119} | {'precision': 0.7665505226480837, 'recall': 0.8262910798122066, 'f1': 0.7953004970628107, 'number': 1065} | 0.7144 | 0.7792 | 0.7454 | 0.8125 |
| 0.3344 | 11.0 | 110 | 0.6505 | {'precision': 0.7031419284940412, 'recall': 0.8022249690976514, 'f1': 0.7494226327944574, 'number': 809} | {'precision': 0.3153153153153153, 'recall': 0.29411764705882354, 'f1': 0.30434782608695654, 'number': 119} | {'precision': 0.755632582322357, 'recall': 0.8187793427230047, 'f1': 0.7859396124380351, 'number': 1065} | 0.7112 | 0.7807 | 0.7443 | 0.8087 |
| 0.3144 | 12.0 | 120 | 0.6461 | {'precision': 0.6973262032085561, 'recall': 0.8059332509270705, 'f1': 0.7477064220183487, 'number': 809} | {'precision': 0.3119266055045872, 'recall': 0.2857142857142857, 'f1': 0.2982456140350877, 'number': 119} | {'precision': 0.7590051457975986, 'recall': 0.8309859154929577, 'f1': 0.7933662034961901, 'number': 1065} | 0.7109 | 0.7883 | 0.7476 | 0.8137 |
| 0.2976 | 13.0 | 130 | 0.6569 | {'precision': 0.6925531914893617, 'recall': 0.8046971569839307, 'f1': 0.7444253859348199, 'number': 809} | {'precision': 0.3025210084033613, 'recall': 0.3025210084033613, 'f1': 0.3025210084033613, 'number': 119} | {'precision': 0.7586805555555556, 'recall': 0.8206572769953052, 'f1': 0.7884528642309426, 'number': 1065} | 0.7060 | 0.7832 | 0.7426 | 0.8094 |
| 0.2876 | 14.0 | 140 | 0.6629 | {'precision': 0.7034632034632035, 'recall': 0.8034610630407911, 'f1': 0.7501442585112521, 'number': 809} | {'precision': 0.3148148148148148, 'recall': 0.2857142857142857, 'f1': 0.29955947136563876, 'number': 119} | {'precision': 0.7657894736842106, 'recall': 0.819718309859155, 'f1': 0.7918367346938776, 'number': 1065} | 0.7169 | 0.7812 | 0.7477 | 0.8104 |
| 0.2877 | 15.0 | 150 | 0.6624 | {'precision': 0.7003222341568206, 'recall': 0.8059332509270705, 'f1': 0.7494252873563217, 'number': 809} | {'precision': 0.3148148148148148, 'recall': 0.2857142857142857, 'f1': 0.29955947136563876, 'number': 119} | {'precision': 0.7602441150828247, 'recall': 0.8187793427230047, 'f1': 0.7884267631103073, 'number': 1065} | 0.7127 | 0.7817 | 0.7456 | 0.8098 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.0
- Tokenizers 0.15.0