README.md

metadata

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 on the funsd dataset. It achieves the following results on the evaluation set:

• Loss: 0.6680
• Answer: {'precision': 0.6523109243697479, 'recall': 0.7676143386897404, 'f1': 0.7052810902896083, 'number': 809}
• Header: {'precision': 0.23300970873786409, 'recall': 0.20168067226890757, 'f1': 0.21621621621621623, 'number': 119}
• Question: {'precision': 0.7324786324786324, 'recall': 0.8046948356807512, 'f1': 0.766890380313199, 'number': 1065}
• Overall Precision: 0.6751
• Overall Recall: 0.7536
• Overall F1: 0.7122
• Overall Accuracy: 0.7957

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

Training results

Training Loss	Epoch	Step	Validation Loss	Answer	Header	Question	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
1.8904	1.0	10	1.6569	{'precision': 0.0226628895184136, 'recall': 0.029666254635352288, 'f1': 0.025695931477516063, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.13233724653148346, 'recall': 0.11643192488262911, 'f1': 0.12387612387612387, 'number': 1065}	0.074	0.0743	0.0741	0.3562
1.5103	2.0	20	1.3215	{'precision': 0.16376306620209058, 'recall': 0.17428924598269468, 'f1': 0.1688622754491018, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.37047970479704795, 'recall': 0.47136150234741786, 'f1': 0.4148760330578512, 'number': 1065}	0.2902	0.3226	0.3055	0.5678
1.1593	3.0	30	0.9985	{'precision': 0.45689655172413796, 'recall': 0.5241038318912238, 'f1': 0.4881980426021877, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.5746268656716418, 'recall': 0.6507042253521127, 'f1': 0.6103038309114928, 'number': 1065}	0.5176	0.5605	0.5382	0.6952
0.8944	4.0	40	0.8291	{'precision': 0.5676982591876208, 'recall': 0.7255871446229913, 'f1': 0.6370048833423766, 'number': 809}	{'precision': 0.125, 'recall': 0.05042016806722689, 'f1': 0.07185628742514971, 'number': 119}	{'precision': 0.648323301805675, 'recall': 0.707981220657277, 'f1': 0.6768402154398564, 'number': 1065}	0.6	0.6759	0.6357	0.7440
0.7344	5.0	50	0.7416	{'precision': 0.6231422505307855, 'recall': 0.7255871446229913, 'f1': 0.670474014848658, 'number': 809}	{'precision': 0.21686746987951808, 'recall': 0.15126050420168066, 'f1': 0.1782178217821782, 'number': 119}	{'precision': 0.6871794871794872, 'recall': 0.7549295774647887, 'f1': 0.7194630872483222, 'number': 1065}	0.6419	0.7070	0.6729	0.7718
0.6312	6.0	60	0.7028	{'precision': 0.616956077630235, 'recall': 0.7466007416563659, 'f1': 0.6756152125279643, 'number': 809}	{'precision': 0.2413793103448276, 'recall': 0.17647058823529413, 'f1': 0.2038834951456311, 'number': 119}	{'precision': 0.7020033388981636, 'recall': 0.7896713615023474, 'f1': 0.7432611577551921, 'number': 1065}	0.6475	0.7356	0.6887	0.7880
0.5603	7.0	70	0.6980	{'precision': 0.6331550802139038, 'recall': 0.7317676143386898, 'f1': 0.6788990825688073, 'number': 809}	{'precision': 0.2604166666666667, 'recall': 0.21008403361344538, 'f1': 0.23255813953488375, 'number': 119}	{'precision': 0.6994219653179191, 'recall': 0.7953051643192488, 'f1': 0.7442882249560634, 'number': 1065}	0.6530	0.7346	0.6914	0.7861
0.5272	8.0	80	0.6733	{'precision': 0.6592827004219409, 'recall': 0.7725587144622992, 'f1': 0.7114399544678428, 'number': 809}	{'precision': 0.25, 'recall': 0.20168067226890757, 'f1': 0.22325581395348837, 'number': 119}	{'precision': 0.7175188600167645, 'recall': 0.8037558685446009, 'f1': 0.758193091231178, 'number': 1065}	0.6728	0.7551	0.7116	0.7927
0.4849	9.0	90	0.6716	{'precision': 0.6549145299145299, 'recall': 0.757725587144623, 'f1': 0.7025787965616046, 'number': 809}	{'precision': 0.23809523809523808, 'recall': 0.21008403361344538, 'f1': 0.22321428571428573, 'number': 119}	{'precision': 0.7216666666666667, 'recall': 0.8131455399061033, 'f1': 0.7646799116997792, 'number': 1065}	0.6711	0.7546	0.7104	0.7961
0.4695	10.0	100	0.6680	{'precision': 0.6523109243697479, 'recall': 0.7676143386897404, 'f1': 0.7052810902896083, 'number': 809}	{'precision': 0.23300970873786409, 'recall': 0.20168067226890757, 'f1': 0.21621621621621623, 'number': 119}	{'precision': 0.7324786324786324, 'recall': 0.8046948356807512, 'f1': 0.766890380313199, 'number': 1065}	0.6751	0.7536	0.7122	0.7957

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.2
• Tokenizers 0.13.3