metadata

tags:
  - generated_from_trainer
datasets:
  - sroie
model-index:
  - name: layoutlm-sroie
    results: []

layoutlm-sroie

This model is a fine-tuned version of microsoft/layoutlm-base-uncased on the sroie dataset. It achieves the following results on the evaluation set:

Loss: 0.0320
Address: {'precision': 0.9154929577464789, 'recall': 0.9365994236311239, 'f1': 0.925925925925926, 'number': 347}
Company: {'precision': 0.9228650137741047, 'recall': 0.9654178674351584, 'f1': 0.9436619718309859, 'number': 347}
Date: {'precision': 0.9884393063583815, 'recall': 0.9855907780979827, 'f1': 0.987012987012987, 'number': 347}
Total: {'precision': 0.8431372549019608, 'recall': 0.8674351585014409, 'f1': 0.8551136363636365, 'number': 347}
Overall Precision: 0.9170
Overall Recall: 0.9388
Overall F1: 0.9277
Overall Accuracy: 0.9942

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

Training results

Training Loss	Epoch	Step	Validation Loss	Address	Company	Date	Total	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.5392	1.0	40	0.1200	{'precision': 0.8346883468834688, 'recall': 0.8876080691642652, 'f1': 0.8603351955307261, 'number': 347}	{'precision': 0.75, 'recall': 0.8126801152737753, 'f1': 0.7800829875518672, 'number': 347}	{'precision': 0.648068669527897, 'recall': 0.8703170028818443, 'f1': 0.7429274292742927, 'number': 347}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 347}	0.7366	0.6427	0.6864	0.9712
0.0803	2.0	80	0.0485	{'precision': 0.896358543417367, 'recall': 0.9221902017291066, 'f1': 0.9090909090909091, 'number': 347}	{'precision': 0.8897849462365591, 'recall': 0.9538904899135446, 'f1': 0.9207232267037552, 'number': 347}	{'precision': 0.9628571428571429, 'recall': 0.9711815561959655, 'f1': 0.9670014347202295, 'number': 347}	{'precision': 0.5219298245614035, 'recall': 0.34293948126801155, 'f1': 0.41391304347826086, 'number': 347}	0.8470	0.7976	0.8215	0.9876
0.0441	3.0	120	0.0373	{'precision': 0.9131652661064426, 'recall': 0.9394812680115274, 'f1': 0.9261363636363636, 'number': 347}	{'precision': 0.9277777777777778, 'recall': 0.962536023054755, 'f1': 0.9448373408769449, 'number': 347}	{'precision': 0.9826589595375722, 'recall': 0.9798270893371758, 'f1': 0.9812409812409811, 'number': 347}	{'precision': 0.5418502202643172, 'recall': 0.7089337175792507, 'f1': 0.6142322097378278, 'number': 347}	0.8214	0.8977	0.8578	0.9897
0.0309	4.0	160	0.0333	{'precision': 0.8864265927977839, 'recall': 0.9221902017291066, 'f1': 0.903954802259887, 'number': 347}	{'precision': 0.915068493150685, 'recall': 0.962536023054755, 'f1': 0.9382022471910113, 'number': 347}	{'precision': 0.9912790697674418, 'recall': 0.9827089337175793, 'f1': 0.9869753979739508, 'number': 347}	{'precision': 0.6556122448979592, 'recall': 0.7406340057636888, 'f1': 0.6955345060893099, 'number': 347}	0.8564	0.9020	0.8786	0.9916
0.0249	5.0	200	0.0318	{'precision': 0.907563025210084, 'recall': 0.9337175792507204, 'f1': 0.9204545454545455, 'number': 347}	{'precision': 0.9054054054054054, 'recall': 0.9654178674351584, 'f1': 0.9344490934449093, 'number': 347}	{'precision': 0.9912536443148688, 'recall': 0.9798270893371758, 'f1': 0.9855072463768115, 'number': 347}	{'precision': 0.7186700767263428, 'recall': 0.8097982708933718, 'f1': 0.7615176151761518, 'number': 347}	0.8761	0.9222	0.8986	0.9924
0.0183	6.0	240	0.0294	{'precision': 0.9103641456582633, 'recall': 0.9365994236311239, 'f1': 0.9232954545454545, 'number': 347}	{'precision': 0.9201101928374655, 'recall': 0.962536023054755, 'f1': 0.9408450704225352, 'number': 347}	{'precision': 0.991304347826087, 'recall': 0.9855907780979827, 'f1': 0.9884393063583815, 'number': 347}	{'precision': 0.8397435897435898, 'recall': 0.7550432276657061, 'f1': 0.795144157814871, 'number': 347}	0.9172	0.9099	0.9136	0.9936
0.0155	7.0	280	0.0290	{'precision': 0.9129213483146067, 'recall': 0.9365994236311239, 'f1': 0.9246088193456615, 'number': 347}	{'precision': 0.9155313351498637, 'recall': 0.968299711815562, 'f1': 0.9411764705882353, 'number': 347}	{'precision': 0.9855907780979827, 'recall': 0.9855907780979827, 'f1': 0.9855907780979827, 'number': 347}	{'precision': 0.8228882833787466, 'recall': 0.8703170028818443, 'f1': 0.8459383753501399, 'number': 347}	0.9081	0.9402	0.9239	0.9940
0.0125	8.0	320	0.0322	{'precision': 0.9103641456582633, 'recall': 0.9365994236311239, 'f1': 0.9232954545454545, 'number': 347}	{'precision': 0.9027027027027027, 'recall': 0.962536023054755, 'f1': 0.9316596931659692, 'number': 347}	{'precision': 0.9884393063583815, 'recall': 0.9855907780979827, 'f1': 0.987012987012987, 'number': 347}	{'precision': 0.8269230769230769, 'recall': 0.8674351585014409, 'f1': 0.8466947960618847, 'number': 347}	0.9061	0.9380	0.9218	0.9936
0.0108	9.0	360	0.0302	{'precision': 0.9078212290502793, 'recall': 0.9365994236311239, 'f1': 0.921985815602837, 'number': 347}	{'precision': 0.9205479452054794, 'recall': 0.968299711815562, 'f1': 0.9438202247191011, 'number': 347}	{'precision': 0.9884393063583815, 'recall': 0.9855907780979827, 'f1': 0.987012987012987, 'number': 347}	{'precision': 0.8386167146974063, 'recall': 0.8386167146974063, 'f1': 0.8386167146974063, 'number': 347}	0.9138	0.9323	0.9230	0.9941
0.0099	10.0	400	0.0304	{'precision': 0.8833333333333333, 'recall': 0.9164265129682997, 'f1': 0.8995756718528995, 'number': 347}	{'precision': 0.9226519337016574, 'recall': 0.962536023054755, 'f1': 0.9421720733427363, 'number': 347}	{'precision': 0.9884393063583815, 'recall': 0.9855907780979827, 'f1': 0.987012987012987, 'number': 347}	{'precision': 0.8472222222222222, 'recall': 0.8789625360230547, 'f1': 0.8628005657708628, 'number': 347}	0.9097	0.9359	0.9226	0.9943
0.0079	11.0	440	0.0312	{'precision': 0.9206798866855525, 'recall': 0.9365994236311239, 'f1': 0.9285714285714285, 'number': 347}	{'precision': 0.9203296703296703, 'recall': 0.9654178674351584, 'f1': 0.9423347398030942, 'number': 347}	{'precision': 0.9884393063583815, 'recall': 0.9855907780979827, 'f1': 0.987012987012987, 'number': 347}	{'precision': 0.8515406162464986, 'recall': 0.8760806916426513, 'f1': 0.8636363636363638, 'number': 347}	0.9197	0.9409	0.9302	0.9945
0.0079	12.0	480	0.0322	{'precision': 0.9101123595505618, 'recall': 0.9337175792507204, 'f1': 0.9217638691322901, 'number': 347}	{'precision': 0.9153005464480874, 'recall': 0.9654178674351584, 'f1': 0.9396914446002805, 'number': 347}	{'precision': 0.9884393063583815, 'recall': 0.9855907780979827, 'f1': 0.987012987012987, 'number': 347}	{'precision': 0.8310626702997275, 'recall': 0.8789625360230547, 'f1': 0.8543417366946778, 'number': 347}	0.9101	0.9409	0.9253	0.9941
0.0068	13.0	520	0.0311	{'precision': 0.9101123595505618, 'recall': 0.9337175792507204, 'f1': 0.9217638691322901, 'number': 347}	{'precision': 0.9226519337016574, 'recall': 0.962536023054755, 'f1': 0.9421720733427363, 'number': 347}	{'precision': 0.991304347826087, 'recall': 0.9855907780979827, 'f1': 0.9884393063583815, 'number': 347}	{'precision': 0.8467966573816156, 'recall': 0.8760806916426513, 'f1': 0.8611898016997168, 'number': 347}	0.9170	0.9395	0.9281	0.9943
0.0068	14.0	560	0.0318	{'precision': 0.9101123595505618, 'recall': 0.9337175792507204, 'f1': 0.9217638691322901, 'number': 347}	{'precision': 0.9201101928374655, 'recall': 0.962536023054755, 'f1': 0.9408450704225352, 'number': 347}	{'precision': 0.9884393063583815, 'recall': 0.9855907780979827, 'f1': 0.987012987012987, 'number': 347}	{'precision': 0.837465564738292, 'recall': 0.8760806916426513, 'f1': 0.8563380281690142, 'number': 347}	0.9132	0.9395	0.9261	0.9943
0.0061	15.0	600	0.0320	{'precision': 0.9154929577464789, 'recall': 0.9365994236311239, 'f1': 0.925925925925926, 'number': 347}	{'precision': 0.9228650137741047, 'recall': 0.9654178674351584, 'f1': 0.9436619718309859, 'number': 347}	{'precision': 0.9884393063583815, 'recall': 0.9855907780979827, 'f1': 0.987012987012987, 'number': 347}	{'precision': 0.8431372549019608, 'recall': 0.8674351585014409, 'f1': 0.8551136363636365, 'number': 347}	0.9170	0.9388	0.9277	0.9942

Framework versions

Transformers 4.28.0
Pytorch 2.1.0+cu118
Datasets 2.15.0
Tokenizers 0.13.3