lilt-en-funsd

someet/lilt-en-funsd

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

29% of all-time downloads

All-time downloads

Public

Parameters

130M

3.6 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors521 MB · 99%

At a glance

Task: Token Classification
Library: transformers
License: mit
Model type: lilt
Access: Public
Created: Feb 28, 2024
Updated: Feb 28, 2024
SHA: ffefe89f

Try a prompt

Base models

SCUT-DLVCLab/lilt-roberta-en-base

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 an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.5254
Answer: {'precision': 0.8486238532110092, 'recall': 0.9057527539779682, 'f1': 0.8762581409117821, 'number': 817}
Header: {'precision': 0.65625, 'recall': 0.5294117647058824, 'f1': 0.586046511627907, 'number': 119}
Question: {'precision': 0.9026629935720845, 'recall': 0.9127205199628597, 'f1': 0.9076638965835643, 'number': 1077}
Overall Precision: 0.8683
Overall Recall: 0.8872
Overall F1: 0.8776
Overall Accuracy: 0.8064

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
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Answer	Header	Question	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.4037	10.53	200	1.0901	{'precision': 0.8236658932714617, 'recall': 0.8690330477356181, 'f1': 0.8457415128052411, 'number': 817}	{'precision': 0.42528735632183906, 'recall': 0.6218487394957983, 'f1': 0.5051194539249146, 'number': 119}	{'precision': 0.871356783919598, 'recall': 0.8050139275766016, 'f1': 0.8368725868725869, 'number': 1077}	0.8129	0.8202	0.8165	0.7725
0.0456	21.05	400	1.4102	{'precision': 0.8165745856353591, 'recall': 0.9045287637698899, 'f1': 0.8583042973286875, 'number': 817}	{'precision': 0.6071428571428571, 'recall': 0.42857142857142855, 'f1': 0.5024630541871921, 'number': 119}	{'precision': 0.8835304822565969, 'recall': 0.9015784586815228, 'f1': 0.8924632352941178, 'number': 1077}	0.8434	0.8748	0.8588	0.7879
0.0146	31.58	600	1.5424	{'precision': 0.834056399132321, 'recall': 0.9412484700122399, 'f1': 0.8844163312248418, 'number': 817}	{'precision': 0.5118110236220472, 'recall': 0.5462184873949579, 'f1': 0.5284552845528455, 'number': 119}	{'precision': 0.9035004730368968, 'recall': 0.8867223769730733, 'f1': 0.895032802249297, 'number': 1077}	0.8495	0.8887	0.8687	0.7913
0.0074	42.11	800	1.4579	{'precision': 0.8571428571428571, 'recall': 0.8886168910648715, 'f1': 0.8725961538461537, 'number': 817}	{'precision': 0.5798319327731093, 'recall': 0.5798319327731093, 'f1': 0.5798319327731093, 'number': 119}	{'precision': 0.8590192644483362, 'recall': 0.9108635097493036, 'f1': 0.8841820639927895, 'number': 1077}	0.8425	0.8823	0.8619	0.8063
0.0043	52.63	1000	1.8745	{'precision': 0.8458100558659218, 'recall': 0.9265605875152999, 'f1': 0.8843457943925235, 'number': 817}	{'precision': 0.5641025641025641, 'recall': 0.5546218487394958, 'f1': 0.559322033898305, 'number': 119}	{'precision': 0.9229268292682927, 'recall': 0.8783658310120706, 'f1': 0.9000951474785919, 'number': 1077}	0.8684	0.8788	0.8736	0.7883
0.0035	63.16	1200	1.8084	{'precision': 0.8344086021505376, 'recall': 0.9498164014687882, 'f1': 0.8883800801373782, 'number': 817}	{'precision': 0.580952380952381, 'recall': 0.5126050420168067, 'f1': 0.5446428571428571, 'number': 119}	{'precision': 0.9076343072573044, 'recall': 0.8941504178272981, 'f1': 0.9008419083255378, 'number': 1077}	0.8588	0.8942	0.8761	0.7965
0.0022	73.68	1400	1.4973	{'precision': 0.8706586826347306, 'recall': 0.8898408812729498, 'f1': 0.8801452784503632, 'number': 817}	{'precision': 0.6176470588235294, 'recall': 0.5294117647058824, 'f1': 0.5701357466063349, 'number': 119}	{'precision': 0.8852313167259787, 'recall': 0.9238625812441968, 'f1': 0.9041344843253067, 'number': 1077}	0.8661	0.8867	0.8763	0.8137
0.0025	84.21	1600	1.5254	{'precision': 0.8486238532110092, 'recall': 0.9057527539779682, 'f1': 0.8762581409117821, 'number': 817}	{'precision': 0.65625, 'recall': 0.5294117647058824, 'f1': 0.586046511627907, 'number': 119}	{'precision': 0.9026629935720845, 'recall': 0.9127205199628597, 'f1': 0.9076638965835643, 'number': 1077}	0.8683	0.8872	0.8776	0.8064
0.0006	94.74	1800	1.5072	{'precision': 0.8583042973286876, 'recall': 0.9045287637698899, 'f1': 0.8808104886769966, 'number': 817}	{'precision': 0.64, 'recall': 0.5378151260504201, 'f1': 0.5844748858447488, 'number': 119}	{'precision': 0.8841354723707665, 'recall': 0.9210770659238626, 'f1': 0.9022282855843565, 'number': 1077}	0.8617	0.8917	0.8765	0.8085
0.0004	105.26	2000	1.5540	{'precision': 0.847926267281106, 'recall': 0.9008567931456548, 'f1': 0.8735905044510385, 'number': 817}	{'precision': 0.5959595959595959, 'recall': 0.4957983193277311, 'f1': 0.5412844036697246, 'number': 119}	{'precision': 0.8814016172506739, 'recall': 0.9108635097493036, 'f1': 0.8958904109589041, 'number': 1077}	0.8538	0.8823	0.8678	0.8014
0.0002	115.79	2200	1.5880	{'precision': 0.8609501738122828, 'recall': 0.9094247246022031, 'f1': 0.8845238095238096, 'number': 817}	{'precision': 0.5876288659793815, 'recall': 0.4789915966386555, 'f1': 0.5277777777777778, 'number': 119}	{'precision': 0.8843416370106761, 'recall': 0.9229340761374187, 'f1': 0.9032258064516129, 'number': 1077}	0.8608	0.8912	0.8758	0.7986
0.0003	126.32	2400	1.5619	{'precision': 0.8586326767091541, 'recall': 0.9069767441860465, 'f1': 0.8821428571428572, 'number': 817}	{'precision': 0.6021505376344086, 'recall': 0.47058823529411764, 'f1': 0.5283018867924528, 'number': 119}	{'precision': 0.8775510204081632, 'recall': 0.9182915506035283, 'f1': 0.8974591651542649, 'number': 1077}	0.8574	0.8872	0.8721	0.8060

Framework versions

Transformers 4.38.1
Pytorch 2.2.1+cu118
Datasets 2.17.1
Tokenizers 0.15.2

Task: Token Classification
Library: transformers
Type: lilt
License: mit
Created: Feb 28, 2024
Updated: Feb 28, 2024