lilt-en-funsd

J10r/lilt-en-funsd

lilt-en-funsd is a token classification model from J10r. 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

28% of all-time downloads

All-time downloads

Public

Parameters

130M

6.8 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: Nov 14, 2024
Updated: Nov 14, 2024
SHA: 5f0f0e55

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: 1.7187
Answer: {'precision': 0.8569767441860465, 'recall': 0.9020807833537332, 'f1': 0.8789505068574837, 'number': 817}
Header: {'precision': 0.6407766990291263, 'recall': 0.5546218487394958, 'f1': 0.5945945945945947, 'number': 119}
Question: {'precision': 0.8962693357597816, 'recall': 0.914577530176416, 'f1': 0.9053308823529412, 'number': 1077}
Overall Precision: 0.8671
Overall Recall: 0.8882
Overall F1: 0.8775
Overall Accuracy: 0.7998

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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
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
0.4061	10.5263	200	1.0439	{'precision': 0.8062770562770563, 'recall': 0.9118727050183598, 'f1': 0.855829982768524, 'number': 817}	{'precision': 0.6043956043956044, 'recall': 0.46218487394957986, 'f1': 0.5238095238095237, 'number': 119}	{'precision': 0.8777173913043478, 'recall': 0.8997214484679665, 'f1': 0.8885832187070151, 'number': 1077}	0.8348	0.8788	0.8562	0.7914
0.0454	21.0526	400	1.4836	{'precision': 0.8243688254665203, 'recall': 0.9192166462668299, 'f1': 0.869212962962963, 'number': 817}	{'precision': 0.48872180451127817, 'recall': 0.5462184873949579, 'f1': 0.5158730158730158, 'number': 119}	{'precision': 0.9097963142580019, 'recall': 0.8709377901578459, 'f1': 0.889943074003795, 'number': 1077}	0.8453	0.8713	0.8581	0.7934
0.0145	31.5789	600	1.3593	{'precision': 0.8672248803827751, 'recall': 0.8873929008567931, 'f1': 0.8771929824561404, 'number': 817}	{'precision': 0.6018518518518519, 'recall': 0.5462184873949579, 'f1': 0.5726872246696034, 'number': 119}	{'precision': 0.8753339269813001, 'recall': 0.9127205199628597, 'f1': 0.8936363636363636, 'number': 1077}	0.8578	0.8808	0.8691	0.8041
0.0074	42.1053	800	1.5465	{'precision': 0.8311111111111111, 'recall': 0.9155446756425949, 'f1': 0.8712871287128713, 'number': 817}	{'precision': 0.6190476190476191, 'recall': 0.5462184873949579, 'f1': 0.5803571428571429, 'number': 119}	{'precision': 0.9031657355679702, 'recall': 0.9006499535747446, 'f1': 0.901906090190609, 'number': 1077}	0.8576	0.8857	0.8715	0.7981
0.0045	52.6316	1000	1.4018	{'precision': 0.8342922899884925, 'recall': 0.8873929008567931, 'f1': 0.860023724792408, 'number': 817}	{'precision': 0.5803571428571429, 'recall': 0.5462184873949579, 'f1': 0.5627705627705628, 'number': 119}	{'precision': 0.8855855855855855, 'recall': 0.9127205199628597, 'f1': 0.8989483310470964, 'number': 1077}	0.8479	0.8808	0.8640	0.8064
0.0031	63.1579	1200	1.6815	{'precision': 0.8714810281517748, 'recall': 0.8714810281517748, 'f1': 0.8714810281517748, 'number': 817}	{'precision': 0.6039603960396039, 'recall': 0.5126050420168067, 'f1': 0.5545454545454545, 'number': 119}	{'precision': 0.8730569948186528, 'recall': 0.9387186629526463, 'f1': 0.9046979865771813, 'number': 1077}	0.8593	0.8862	0.8726	0.7952
0.0018	73.6842	1400	1.5823	{'precision': 0.8553530751708428, 'recall': 0.9192166462668299, 'f1': 0.8861356932153391, 'number': 817}	{'precision': 0.5865384615384616, 'recall': 0.5126050420168067, 'f1': 0.5470852017937219, 'number': 119}	{'precision': 0.8930530164533821, 'recall': 0.9071494893221913, 'f1': 0.9000460617227084, 'number': 1077}	0.8618	0.8887	0.8750	0.8061
0.0015	84.2105	1600	1.6540	{'precision': 0.8509895227008148, 'recall': 0.8947368421052632, 'f1': 0.8723150357995225, 'number': 817}	{'precision': 0.6477272727272727, 'recall': 0.4789915966386555, 'f1': 0.5507246376811594, 'number': 119}	{'precision': 0.8776408450704225, 'recall': 0.9257195914577531, 'f1': 0.9010393131495708, 'number': 1077}	0.8569	0.8867	0.8716	0.8039
0.0005	94.7368	1800	1.7397	{'precision': 0.8578199052132701, 'recall': 0.8861689106487148, 'f1': 0.8717639975918122, 'number': 817}	{'precision': 0.5740740740740741, 'recall': 0.5210084033613446, 'f1': 0.5462555066079295, 'number': 119}	{'precision': 0.8785971223021583, 'recall': 0.9071494893221913, 'f1': 0.8926450433988122, 'number': 1077}	0.8542	0.8758	0.8649	0.7925
0.0003	105.2632	2000	1.6680	{'precision': 0.8688915375446961, 'recall': 0.8922888616891065, 'f1': 0.8804347826086957, 'number': 817}	{'precision': 0.6122448979591837, 'recall': 0.5042016806722689, 'f1': 0.5529953917050692, 'number': 119}	{'precision': 0.8774250440917107, 'recall': 0.9238625812441968, 'f1': 0.9000452284034374, 'number': 1077}	0.8614	0.8862	0.8737	0.8011
0.0002	115.7895	2200	1.6812	{'precision': 0.8494252873563218, 'recall': 0.9045287637698899, 'f1': 0.8761114404267932, 'number': 817}	{'precision': 0.6704545454545454, 'recall': 0.4957983193277311, 'f1': 0.5700483091787439, 'number': 119}	{'precision': 0.8914798206278027, 'recall': 0.9229340761374187, 'f1': 0.906934306569343, 'number': 1077}	0.8644	0.8902	0.8771	0.8051
0.0004	126.3158	2400	1.7187	{'precision': 0.8569767441860465, 'recall': 0.9020807833537332, 'f1': 0.8789505068574837, 'number': 817}	{'precision': 0.6407766990291263, 'recall': 0.5546218487394958, 'f1': 0.5945945945945947, 'number': 119}	{'precision': 0.8962693357597816, 'recall': 0.914577530176416, 'f1': 0.9053308823529412, 'number': 1077}	0.8671	0.8882	0.8775	0.7998

Framework versions

Transformers 4.46.2
Pytorch 2.5.0+cu121
Datasets 3.1.0
Tokenizers 0.20.3

Task: Token Classification
Library: transformers
Type: lilt
License: mit
Created: Nov 14, 2024
Updated: Nov 14, 2024