layoutlm-funsd

elvinaqa/layoutlm-funsd

layoutlm-funsd is a token classification model from elvinaqa. Use it when you need labels on individual words, such as names. It is set up for transformers.

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

Downloads · 30 days

7% of all-time downloads

All-time downloads

349

Public

Parameters

113M

1.4 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin451 MB · 50%

Parameter types

How the weights are stored.

F32113M · 100%

Try a prompt

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: 0.7158
Answer: {'precision': 0.7149220489977728, 'recall': 0.7935723114956736, 'f1': 0.7521968365553603, 'number': 809}
Header: {'precision': 0.3063063063063063, 'recall': 0.2857142857142857, 'f1': 0.2956521739130435, 'number': 119}
Question: {'precision': 0.7892857142857143, 'recall': 0.8300469483568075, 'f1': 0.8091533180778031, 'number': 1065}
Overall Precision: 0.7327
Overall Recall: 0.7827
Overall F1: 0.7569
Overall Accuracy: 0.8108

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
1.8132	1.0	10	1.6191	{'precision': 0.015122873345935728, 'recall': 0.019777503090234856, 'f1': 0.01713979646491698, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.1685508735868448, 'recall': 0.1539906103286385, 'f1': 0.16094210009813542, 'number': 1065}	0.0886	0.0903	0.0895	0.3534
1.4783	2.0	20	1.2483	{'precision': 0.12857142857142856, 'recall': 0.12237330037082818, 'f1': 0.1253958201393287, 'number': 809}	{'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 119}	{'precision': 0.4541955350269438, 'recall': 0.5539906103286385, 'f1': 0.4991539763113368, 'number': 1065}	0.3330	0.3457	0.3392	0.5682
1.1072	3.0	30	0.9718	{'precision': 0.42777155655095184, 'recall': 0.4721878862793572, 'f1': 0.4488836662749706, 'number': 809}	{'precision': 0.04, 'recall': 0.008403361344537815, 'f1': 0.01388888888888889, 'number': 119}	{'precision': 0.6266205704407951, 'recall': 0.6807511737089202, 'f1': 0.6525652565256526, 'number': 1065}	0.5340	0.5559	0.5447	0.7070
0.8444	4.0	40	0.7957	{'precision': 0.6296296296296297, 'recall': 0.7354758961681088, 'f1': 0.6784492588369442, 'number': 809}	{'precision': 0.19230769230769232, 'recall': 0.08403361344537816, 'f1': 0.11695906432748539, 'number': 119}	{'precision': 0.6831168831168831, 'recall': 0.7408450704225352, 'f1': 0.7108108108108109, 'number': 1065}	0.6478	0.6994	0.6726	0.7651
0.6845	5.0	50	0.7443	{'precision': 0.6530612244897959, 'recall': 0.7515451174289246, 'f1': 0.6988505747126437, 'number': 809}	{'precision': 0.23684210526315788, 'recall': 0.15126050420168066, 'f1': 0.1846153846153846, 'number': 119}	{'precision': 0.7318181818181818, 'recall': 0.755868544600939, 'f1': 0.74364896073903, 'number': 1065}	0.6792	0.7180	0.6980	0.7736
0.5597	6.0	60	0.6918	{'precision': 0.6673706441393875, 'recall': 0.7812113720642769, 'f1': 0.7198177676537586, 'number': 809}	{'precision': 0.2857142857142857, 'recall': 0.15126050420168066, 'f1': 0.1978021978021978, 'number': 119}	{'precision': 0.7344150298889838, 'recall': 0.8075117370892019, 'f1': 0.7692307692307693, 'number': 1065}	0.6923	0.7577	0.7235	0.7933
0.4929	7.0	70	0.6803	{'precision': 0.6694825765575502, 'recall': 0.7836835599505563, 'f1': 0.7220956719817767, 'number': 809}	{'precision': 0.21818181818181817, 'recall': 0.20168067226890757, 'f1': 0.2096069868995633, 'number': 119}	{'precision': 0.7467134092900964, 'recall': 0.8, 'f1': 0.772438803263826, 'number': 1065}	0.6870	0.7577	0.7206	0.7945
0.4447	8.0	80	0.6814	{'precision': 0.6866158868335147, 'recall': 0.7799752781211372, 'f1': 0.7303240740740741, 'number': 809}	{'precision': 0.26506024096385544, 'recall': 0.18487394957983194, 'f1': 0.21782178217821785, 'number': 119}	{'precision': 0.7810283687943262, 'recall': 0.8272300469483568, 'f1': 0.8034655722754217, 'number': 1065}	0.7202	0.7697	0.7441	0.8024
0.3953	9.0	90	0.6739	{'precision': 0.7015765765765766, 'recall': 0.7700865265760197, 'f1': 0.7342368886269888, 'number': 809}	{'precision': 0.2920353982300885, 'recall': 0.2773109243697479, 'f1': 0.28448275862068967, 'number': 119}	{'precision': 0.7753496503496503, 'recall': 0.8328638497652582, 'f1': 0.8030783159800814, 'number': 1065}	0.7193	0.7742	0.7458	0.8115
0.3538	10.0	100	0.6853	{'precision': 0.7081497797356828, 'recall': 0.7948084054388134, 'f1': 0.7489807804309844, 'number': 809}	{'precision': 0.32673267326732675, 'recall': 0.2773109243697479, 'f1': 0.30000000000000004, 'number': 119}	{'precision': 0.7804878048780488, 'recall': 0.8413145539906103, 'f1': 0.8097605061003164, 'number': 1065}	0.7288	0.7888	0.7576	0.8152
0.3262	11.0	110	0.6948	{'precision': 0.7058177826564215, 'recall': 0.7948084054388134, 'f1': 0.7476744186046511, 'number': 809}	{'precision': 0.35051546391752575, 'recall': 0.2857142857142857, 'f1': 0.3148148148148148, 'number': 119}	{'precision': 0.8032638259292838, 'recall': 0.831924882629108, 'f1': 0.8173431734317343, 'number': 1065}	0.7404	0.7842	0.7617	0.8129
0.3094	12.0	120	0.6989	{'precision': 0.7128603104212861, 'recall': 0.7948084054388134, 'f1': 0.7516072472238456, 'number': 809}	{'precision': 0.3333333333333333, 'recall': 0.29411764705882354, 'f1': 0.3125, 'number': 119}	{'precision': 0.7987364620938628, 'recall': 0.8309859154929577, 'f1': 0.8145421076852278, 'number': 1065}	0.7390	0.7842	0.7610	0.8138
0.2941	13.0	130	0.7134	{'precision': 0.7239819004524887, 'recall': 0.7911001236093943, 'f1': 0.7560543414057885, 'number': 809}	{'precision': 0.32710280373831774, 'recall': 0.29411764705882354, 'f1': 0.3097345132743363, 'number': 119}	{'precision': 0.7998204667863554, 'recall': 0.8366197183098592, 'f1': 0.8178063331803579, 'number': 1065}	0.7439	0.7858	0.7643	0.8115
0.2813	14.0	140	0.7138	{'precision': 0.7106710671067107, 'recall': 0.7985166872682324, 'f1': 0.7520372526193247, 'number': 809}	{'precision': 0.3119266055045872, 'recall': 0.2857142857142857, 'f1': 0.2982456140350877, 'number': 119}	{'precision': 0.7935656836461126, 'recall': 0.8338028169014085, 'f1': 0.8131868131868133, 'number': 1065}	0.7337	0.7868	0.7593	0.8109
0.2812	15.0	150	0.7158	{'precision': 0.7149220489977728, 'recall': 0.7935723114956736, 'f1': 0.7521968365553603, 'number': 809}	{'precision': 0.3063063063063063, 'recall': 0.2857142857142857, 'f1': 0.2956521739130435, 'number': 119}	{'precision': 0.7892857142857143, 'recall': 0.8300469483568075, 'f1': 0.8091533180778031, 'number': 1065}	0.7327	0.7827	0.7569	0.8108

Framework versions

Transformers 4.24.0
Pytorch 1.12.1+cu113
Datasets 2.6.1
Tokenizers 0.13.1

Task: Token Classification
Library: transformers
Type: layoutlm
Created: Nov 4, 2022
Updated: Jul 2, 2023