layoutlm-funsd

Neha-CanWill/layoutlm-funsd

layoutlm-funsd is a token classification model from Neha-CanWill. Use it when you need labels on individual words, such as names. It is set up for transformers. The card lists the license as mit.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

11% of all-time downloads

All-time downloads

Public

Parameters

113M

15.3 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors451 MB · 100%

At a glance

Task: Token Classification
Library: transformers
License: mit
Model type: layoutlm
Access: Public
Created: Mar 8, 2024
Updated: Mar 11, 2024
SHA: 35eb996f

Try a prompt

Base models

microsoft/layoutlm-base-uncased

Datasets

funsd

Model card

From the Hugging Face model README

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: 1.1246
Answer: {'precision': 0.3804878048780488, 'recall': 0.4820766378244747, 'f1': 0.425299890948746, 'number': 809}
Header: {'precision': 0.34408602150537637, 'recall': 0.2689075630252101, 'f1': 0.3018867924528302, 'number': 119}
Question: {'precision': 0.4845360824742268, 'recall': 0.6178403755868545, 'f1': 0.5431283532810565, 'number': 1065}
Overall Precision: 0.4362
Overall Recall: 0.5419
Overall F1: 0.4833
Overall Accuracy: 0.6171

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	Answer	Header	Question	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
1.7202	1.0	10	1.4980	{'precision': 0.05310734463276836, 'recall': 0.0580964153275649, 'f1': 0.05548996458087367, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.26246719160104987, 'recall': 0.28169014084507044, 'f1': 0.27173913043478265, 'number': 1065}	0.1711	0.1741	0.1726	0.3625
1.4151	2.0	20	1.3029	{'precision': 0.19834183673469388, 'recall': 0.38442521631644005, 'f1': 0.26167437946992006, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.266388557806913, 'recall': 0.4197183098591549, 'f1': 0.32592052497265767, 'number': 1065}	0.2325	0.3803	0.2886	0.4280
1.259	3.0	30	1.1884	{'precision': 0.2627235213204952, 'recall': 0.4721878862793572, 'f1': 0.3376049491825011, 'number': 809}	{'precision': 0.06349206349206349, 'recall': 0.03361344537815126, 'f1': 0.04395604395604396, 'number': 119}	{'precision': 0.3270588235294118, 'recall': 0.5220657276995305, 'f1': 0.4021699819168174, 'number': 1065}	0.2928	0.4727	0.3616	0.4939
1.1328	4.0	40	1.0951	{'precision': 0.30996309963099633, 'recall': 0.519159456118665, 'f1': 0.3881700554528651, 'number': 809}	{'precision': 0.2857142857142857, 'recall': 0.18487394957983194, 'f1': 0.22448979591836735, 'number': 119}	{'precision': 0.4103139013452915, 'recall': 0.5154929577464789, 'f1': 0.4569288389513109, 'number': 1065}	0.3578	0.4972	0.4161	0.5748
1.0223	5.0	50	1.0810	{'precision': 0.28736581337737405, 'recall': 0.43016069221260816, 'f1': 0.3445544554455445, 'number': 809}	{'precision': 0.37142857142857144, 'recall': 0.2184873949579832, 'f1': 0.2751322751322751, 'number': 119}	{'precision': 0.38396624472573837, 'recall': 0.5981220657276995, 'f1': 0.4676945668135095, 'number': 1065}	0.3439	0.5073	0.4099	0.5856
0.9408	6.0	60	1.0602	{'precision': 0.3160667251975417, 'recall': 0.44499381953028433, 'f1': 0.3696098562628337, 'number': 809}	{'precision': 0.345679012345679, 'recall': 0.23529411764705882, 'f1': 0.27999999999999997, 'number': 119}	{'precision': 0.4154838709677419, 'recall': 0.6046948356807512, 'f1': 0.49254302103250486, 'number': 1065}	0.3726	0.5178	0.4333	0.5983
0.8629	7.0	70	1.0853	{'precision': 0.3160220994475138, 'recall': 0.3535228677379481, 'f1': 0.33372228704784135, 'number': 809}	{'precision': 0.375, 'recall': 0.2773109243697479, 'f1': 0.31884057971014496, 'number': 119}	{'precision': 0.42748091603053434, 'recall': 0.6309859154929578, 'f1': 0.50967007963595, 'number': 1065}	0.3864	0.4972	0.4348	0.5961
0.8089	8.0	80	1.0864	{'precision': 0.35083114610673666, 'recall': 0.4956736711990111, 'f1': 0.4108606557377049, 'number': 809}	{'precision': 0.36904761904761907, 'recall': 0.2605042016806723, 'f1': 0.30541871921182273, 'number': 119}	{'precision': 0.4398051496172582, 'recall': 0.5934272300469483, 'f1': 0.5051958433253397, 'number': 1065}	0.3994	0.5339	0.4569	0.6110
0.7662	9.0	90	1.0967	{'precision': 0.36006974716652135, 'recall': 0.5105067985166872, 'f1': 0.42229038854805717, 'number': 809}	{'precision': 0.4266666666666667, 'recall': 0.2689075630252101, 'f1': 0.32989690721649484, 'number': 119}	{'precision': 0.4724770642201835, 'recall': 0.5802816901408451, 'f1': 0.5208596713021492, 'number': 1065}	0.4202	0.5334	0.4700	0.6115
0.7718	10.0	100	1.1450	{'precision': 0.375, 'recall': 0.5414091470951793, 'f1': 0.44309559939301973, 'number': 809}	{'precision': 0.4050632911392405, 'recall': 0.2689075630252101, 'f1': 0.3232323232323232, 'number': 119}	{'precision': 0.5078125, 'recall': 0.5492957746478874, 'f1': 0.5277401894451962, 'number': 1065}	0.4398	0.5294	0.4804	0.6057
0.6988	11.0	110	1.1180	{'precision': 0.36609829488465395, 'recall': 0.4511742892459827, 'f1': 0.4042081949058693, 'number': 809}	{'precision': 0.3333333333333333, 'recall': 0.2689075630252101, 'f1': 0.29767441860465116, 'number': 119}	{'precision': 0.4661602209944751, 'recall': 0.6338028169014085, 'f1': 0.5372065260644648, 'number': 1065}	0.4219	0.5379	0.4729	0.6089
0.6905	12.0	120	1.1064	{'precision': 0.36837029893924783, 'recall': 0.4721878862793572, 'f1': 0.41386782231852653, 'number': 809}	{'precision': 0.3793103448275862, 'recall': 0.2773109243697479, 'f1': 0.32038834951456313, 'number': 119}	{'precision': 0.47112676056338026, 'recall': 0.6281690140845071, 'f1': 0.5384305835010061, 'number': 1065}	0.4261	0.5439	0.4778	0.6149
0.666	13.0	130	1.1045	{'precision': 0.36981132075471695, 'recall': 0.484548825710754, 'f1': 0.4194756554307116, 'number': 809}	{'precision': 0.3516483516483517, 'recall': 0.2689075630252101, 'f1': 0.3047619047619048, 'number': 119}	{'precision': 0.48205128205128206, 'recall': 0.6178403755868545, 'f1': 0.5415637860082304, 'number': 1065}	0.4300	0.5429	0.4799	0.6174
0.6335	14.0	140	1.1195	{'precision': 0.3810463968410661, 'recall': 0.47713226205191595, 'f1': 0.42371020856201974, 'number': 809}	{'precision': 0.34831460674157305, 'recall': 0.2605042016806723, 'f1': 0.2980769230769231, 'number': 119}	{'precision': 0.4817204301075269, 'recall': 0.6309859154929578, 'f1': 0.5463414634146342, 'number': 1065}	0.4361	0.5464	0.4851	0.6187
0.6277	15.0	150	1.1246	{'precision': 0.3804878048780488, 'recall': 0.4820766378244747, 'f1': 0.425299890948746, 'number': 809}	{'precision': 0.34408602150537637, 'recall': 0.2689075630252101, 'f1': 0.3018867924528302, 'number': 119}	{'precision': 0.4845360824742268, 'recall': 0.6178403755868545, 'f1': 0.5431283532810565, 'number': 1065}	0.4362	0.5419	0.4833	0.6171

Framework versions

Transformers 4.38.2
Pytorch 2.2.1+cu121
Datasets 2.18.0
Tokenizers 0.15.2

Task: Token Classification
Library: transformers
Type: layoutlm
License: mit
Created: Mar 8, 2024
Updated: Mar 11, 2024