lilt-en-funsd

AHMED36/lilt-en-funsd

lilt-en-funsd is a token classification model from AHMED36. 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

Repo size

4.2 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin521 MB · 99%

At a glance

Task: Token Classification
Library: transformers
License: mit
Model type: lilt
Access: Public
Created: Sep 19, 2023
Updated: Sep 19, 2023
SHA: f1fa33a8

Try a prompt

Base models

SCUT-DLVCLab/lilt-roberta-en-base

Datasets

funsd-layoutlmv3

Model card

From the Hugging Face model README

lilt-en-funsd

This model is a fine-tuned version of SCUT-DLVCLab/lilt-roberta-en-base on the funsd-layoutlmv3 dataset. It achieves the following results on the evaluation set:

Loss: 2.9565
Answer: {'precision': 0.8948004836759371, 'recall': 0.9057527539779682, 'f1': 0.9002433090024331, 'number': 817}
Header: {'precision': 0.6868686868686869, 'recall': 0.5714285714285714, 'f1': 0.6238532110091742, 'number': 119}
Question: {'precision': 0.8923212709620476, 'recall': 0.9387186629526463, 'f1': 0.9149321266968325, 'number': 1077}
Overall Precision: 0.8834
Overall Recall: 0.9036
Overall F1: 0.8934
Overall Accuracy: 0.8096

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: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
training_steps: 2500

Training results

Training Loss	Epoch	Step	Validation Loss	Answer	Header	Question	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.409	10.53	200	0.8991	{'precision': 0.8176855895196506, 'recall': 0.9167686658506732, 'f1': 0.8643969994229659, 'number': 817}	{'precision': 0.5094339622641509, 'recall': 0.453781512605042, 'f1': 0.48, 'number': 119}	{'precision': 0.891465677179963, 'recall': 0.8922934076137419, 'f1': 0.8918793503480278, 'number': 1077}	0.84	0.8763	0.8578	0.7897
0.0485	21.05	400	1.1875	{'precision': 0.8504566210045662, 'recall': 0.9118727050183598, 'f1': 0.8800945067926758, 'number': 817}	{'precision': 0.5691056910569106, 'recall': 0.5882352941176471, 'f1': 0.578512396694215, 'number': 119}	{'precision': 0.8970315398886828, 'recall': 0.8978644382544104, 'f1': 0.897447795823666, 'number': 1077}	0.8580	0.8852	0.8714	0.7935
0.0139	31.58	600	1.5032	{'precision': 0.8455377574370709, 'recall': 0.9045287637698899, 'f1': 0.8740390301596689, 'number': 817}	{'precision': 0.6206896551724138, 'recall': 0.6050420168067226, 'f1': 0.6127659574468085, 'number': 119}	{'precision': 0.9057142857142857, 'recall': 0.883008356545961, 'f1': 0.8942172073342736, 'number': 1077}	0.8637	0.8753	0.8695	0.7913
0.0083	42.11	800	1.4968	{'precision': 0.8316939890710382, 'recall': 0.9314565483476133, 'f1': 0.8787528868360277, 'number': 817}	{'precision': 0.6363636363636364, 'recall': 0.47058823529411764, 'f1': 0.5410628019323671, 'number': 119}	{'precision': 0.8928909952606635, 'recall': 0.8746518105849582, 'f1': 0.8836772983114447, 'number': 1077}	0.8547	0.8738	0.8642	0.8017
0.0058	52.63	1000	1.7837	{'precision': 0.8385300668151447, 'recall': 0.9216646266829865, 'f1': 0.8781341107871721, 'number': 817}	{'precision': 0.6138613861386139, 'recall': 0.5210084033613446, 'f1': 0.5636363636363637, 'number': 119}	{'precision': 0.8972667295004713, 'recall': 0.8839368616527391, 'f1': 0.8905519176800748, 'number': 1077}	0.8578	0.8778	0.8677	0.7914
0.008	63.16	1200	1.8600	{'precision': 0.8239130434782609, 'recall': 0.9277845777233782, 'f1': 0.8727691421991941, 'number': 817}	{'precision': 0.5865384615384616, 'recall': 0.5126050420168067, 'f1': 0.5470852017937219, 'number': 119}	{'precision': 0.9037735849056604, 'recall': 0.8895078922934077, 'f1': 0.8965839962564343, 'number': 1077}	0.8527	0.8828	0.8675	0.8009
0.0037	73.68	1400	2.8372	{'precision': 0.8821428571428571, 'recall': 0.9069767441860465, 'f1': 0.8943874471937237, 'number': 817}	{'precision': 0.5966386554621849, 'recall': 0.5966386554621849, 'f1': 0.5966386554621849, 'number': 119}	{'precision': 0.8961748633879781, 'recall': 0.9136490250696379, 'f1': 0.9048275862068965, 'number': 1077}	0.8731	0.8922	0.8826	0.7928
0.004	84.21	1600	2.8378	{'precision': 0.881578947368421, 'recall': 0.9020807833537332, 'f1': 0.8917120387174834, 'number': 817}	{'precision': 0.631578947368421, 'recall': 0.6050420168067226, 'f1': 0.6180257510729613, 'number': 119}	{'precision': 0.891989198919892, 'recall': 0.9201485608170845, 'f1': 0.9058500914076782, 'number': 1077}	0.8734	0.8942	0.8837	0.8079
0.0018	94.74	1800	3.0272	{'precision': 0.8742655699177438, 'recall': 0.9106487148102815, 'f1': 0.8920863309352519, 'number': 817}	{'precision': 0.6759259259259259, 'recall': 0.6134453781512605, 'f1': 0.6431718061674008, 'number': 119}	{'precision': 0.89937106918239, 'recall': 0.9294336118848654, 'f1': 0.9141552511415526, 'number': 1077}	0.8774	0.9031	0.8901	0.7992
0.0008	105.26	2000	2.9565	{'precision': 0.8948004836759371, 'recall': 0.9057527539779682, 'f1': 0.9002433090024331, 'number': 817}	{'precision': 0.6868686868686869, 'recall': 0.5714285714285714, 'f1': 0.6238532110091742, 'number': 119}	{'precision': 0.8923212709620476, 'recall': 0.9387186629526463, 'f1': 0.9149321266968325, 'number': 1077}	0.8834	0.9036	0.8934	0.8096
0.0008	115.79	2200	3.1429	{'precision': 0.8411111111111111, 'recall': 0.9265605875152999, 'f1': 0.881770529994176, 'number': 817}	{'precision': 0.6666666666666666, 'recall': 0.5546218487394958, 'f1': 0.6055045871559633, 'number': 119}	{'precision': 0.9147141518275539, 'recall': 0.9062209842154132, 'f1': 0.9104477611940299, 'number': 1077}	0.8708	0.8937	0.8821	0.7970
0.0005	126.32	2400	3.0269	{'precision': 0.8617511520737328, 'recall': 0.9155446756425949, 'f1': 0.8878338278931751, 'number': 817}	{'precision': 0.6952380952380952, 'recall': 0.6134453781512605, 'f1': 0.6517857142857143, 'number': 119}	{'precision': 0.906871609403255, 'recall': 0.9312906220984215, 'f1': 0.9189189189189189, 'number': 1077}	0.8773	0.9061	0.8915	0.7994

Framework versions

Transformers 4.32.0
Pytorch 2.0.1+cu117
Datasets 2.13.1
Tokenizers 0.13.3

Task: Token Classification
Library: transformers
Type: lilt
License: mit
Created: Sep 19, 2023
Updated: Sep 19, 2023