This is a SpanMarker model trained on the FewNERD dataset that can be used for Named Entity Recognition. This SpanMarker model uses numind/generic-entity_recognition_NER-v1 as the underlying encoder.

Model Details

Model Description

Model Type: SpanMarker
Encoder: numind/generic-entity_recognition_NER-v1
Maximum Sequence Length: 256 tokens
Maximum Entity Length: 8 words
Training Dataset: FewNERD
Language: en
License: cc-by-sa-4.0

Model Sources

Repository: SpanMarker on GitHub
Thesis: SpanMarker For Named Entity Recognition

Model Labels

Label	Examples
art-broadcastprogram	"Corazones", "The Gale Storm Show : Oh , Susanna", "Street Cents"
art-film	"Shawshank Redemption", "L'Atlantide", "Bosch"
art-music	"Hollywood Studio Symphony", "Atkinson , Danko and Ford ( with Brockie and Hilton )", "Champion Lover"
art-other	"The Today Show", "Venus de Milo", "Aphrodite of Milos"
art-painting	"Production/Reproduction", "Touit", "Cofiwch Dryweryn"
art-writtenart	"The Seven Year Itch", "Imelda de ' Lambertazzi", "Time"
building-airport	"Sheremetyevo International Airport", "Newark Liberty International Airport", "Luton Airport"
building-hospital	"Yeungnam University Hospital", "Hokkaido University Hospital", "Memorial Sloan-Kettering Cancer Center"
building-hotel	"The Standard Hotel", "Flamingo Hotel", "Radisson Blu Sea Plaza Hotel"
building-library	"British Library", "Bayerische Staatsbibliothek", "Berlin State Library"
building-other	"Henry Ford Museum", "Alpha Recording Studios", "Communiplex"
building-restaurant	"Carnegie Deli", "Fatburger", "Trumbull"
building-sportsfacility	"Boston Garden", "Sports Center", "Glenn Warner Soccer Facility"
building-theater	"Sanders Theatre", "National Paris Opera", "Pittsburgh Civic Light Opera"
event-attack/battle/war/militaryconflict	"Easter Offensive", "Jurist", "Vietnam War"
event-disaster	"the 1912 North Mount Lyell Disaster", "1990s North Korean famine", "1693 Sicily earthquake"
event-election	"Elections to the European Parliament", "March 1898 elections", "1982 Mitcham and Morden by-election"
event-other	"Union for a Popular Movement", "Masaryk Democratic Movement", "Eastwood Scoring Stage"
event-protest	"Iranian Constitutional Revolution", "French Revolution", "Russian Revolution"
event-sportsevent	"World Cup", "National Champions", "Stanley Cup"
location-GPE	"Croatian", "Mediterranean Basin", "the Republic of Croatia"
location-bodiesofwater	"Arthur Kill", "Atatürk Dam Lake", "Norfolk coast"
location-island	"new Samsat district", "Laccadives", "Staten Island"
location-mountain	"Salamander Glacier", "Miteirya Ridge", "Ruweisat Ridge"
location-other	"Victoria line", "Northern City Line", "Cartuther"
location-park	"Painted Desert Community Complex Historic District", "Gramercy Park", "Shenandoah National Park"
location-road/railway/highway/transit	"NJT", "Newark-Elizabeth Rail Link", "Friern Barnet Road"
organization-company	"Texas Chicken", "Dixy Chicken", "Church 's Chicken"
organization-education	"MIT", "Belfast Royal Academy and the Ulster College of Physical Education", "Barnard College"
organization-government/governmentagency	"Congregazione dei Nobili", "Diet", "Supreme Court"
organization-media/newspaper	"Clash", "Al Jazeera", "TimeOut Melbourne"
organization-other	"Defence Sector C", "IAEA", "4th Army"
organization-politicalparty	"Al Wafa ' Islamic", "Shimpotō", "Kenseitō"
organization-religion	"UPCUSA", "Christian", "Jewish"
organization-showorganization	"Lizzy", "Bochumer Symphoniker", "Mr. Mister"
organization-sportsleague	"China League One", "NHL", "First Division"
organization-sportsteam	"Arsenal", "Luc Alphand Aventures", "Tottenham"
other-astronomything	"Algol", "`` Caput Larvae ''", "Zodiac"
other-award	"Order of the Republic of Guinea and Nigeria", "Grand Commander of the Order of the Niger", "GCON"
other-biologything	"N-terminal lipid", "Amphiphysin", "BAR"
other-chemicalthing	"uranium", "carbon dioxide", "sulfur"
other-currency	"$", "lac crore", "Travancore Rupee"
other-disease	"bladder cancer", "French Dysentery Epidemic of 1779", "hypothyroidism"
other-educationaldegree	"BSc ( Hons ) in physics", "Bachelor", "Master"
other-god	"Raijin", "Fujin", "El"
other-language	"Breton-speaking", "Latin", "English"
other-law	"Leahy–Smith America Invents Act ( AIA", "United States Freedom Support Act", "Thirty Years ' Peace"
other-livingthing	"monkeys", "patchouli", "insects"
other-medical	"amitriptyline", "Pediatrics", "pediatrician"
person-actor	"Tchéky Karyo", "Edmund Payne", "Ellaline Terriss"
person-artist/author	"Hicks", "Gaetano Donizett", "George Axelrod"
person-athlete	"Tozawa", "Neville", "Jaguar"
person-director	"Richard Quine", "Bob Swaim", "Frank Darabont"
person-other	"Campbell", "Holden", "Richard Benson"
person-politician	"William", "Rivière", "Emeric"
person-scholar	"Wurdack", "Stalmine", "Stedman"
person-soldier	"Joachim Ziegler", "Helmuth Weidling", "Krukenberg"
product-airplane	"Spey-equipped FGR.2s", "EC135T2 CPDS", "Luton"
product-car	"Phantom", "100EX", "Corvettes - GT1 C6R"
product-food	"red grape", "yakiniku", "V. labrusca"
product-game	"Hardcore RPG", "Splinter Cell", "Airforce Delta"
product-other	"X11", "PDP-1", "Fairbottom Bobs"
product-ship	"Essex", "Congress", "HMS `` Chinkara ''"
product-software	"AmiPDF", "Wikipedia", "Apdf"
product-train	"55022", "Royal Scots Grey", "High Speed Trains"
product-weapon	"AR-15 's", "ZU-23-2MR Wróbel II", "ZU-23-2M Wróbel"

Uses

Direct Use for Inference

python

from span_marker import SpanMarkerModel

# Download from the 🤗 Hub
model = SpanMarkerModel.from_pretrained("guishe/span-marker-generic-ner-v1-fewnerd-fine-super")
# Run inference
entities = model.predict("Most of the Steven Seagal movie \"Under Siege \"(co-starring Tommy Lee Jones) was filmed on the, which is docked on Mobile Bay at Battleship Memorial Park and open to the public.")

Downstream Use

You can finetune this model on your own dataset.

Click to expand

python

from span_marker import SpanMarkerModel, Trainer

# Download from the 🤗 Hub
model = SpanMarkerModel.from_pretrained("guishe/span-marker-generic-ner-v1-fewnerd-fine-super")

# Specify a Dataset with "tokens" and "ner_tag" columns
dataset = load_dataset("conll2003") # For example CoNLL2003

# Initialize a Trainer using the pretrained model & dataset
trainer = Trainer(
    model=model,
    train_dataset=dataset["train"],
    eval_dataset=dataset["validation"],
)
trainer.train()
trainer.save_model("guishe/span-marker-generic-ner-v1-fewnerd-fine-super-finetuned")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Sentence length	1	24.4945	267
Entities per sentence	0	2.5832	88

Training Hyperparameters

learning_rate: 1e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 3

Training Results

Epoch	Step	Validation Loss	Validation Precision	Validation Recall	Validation F1	Validation Accuracy
0.2980	3000	0.0290	0.6503	0.6402	0.6452	0.9109
0.5961	6000	0.0250	0.6749	0.6794	0.6772	0.9202
0.8941	9000	0.0236	0.6908	0.6871	0.6889	0.9229
1.1921	12000	0.0234	0.6853	0.7007	0.6929	0.9239
1.4902	15000	0.0227	0.6966	0.6929	0.6948	0.9241
1.7882	18000	0.0221	0.7073	0.6922	0.6997	0.9250
2.0862	21000	0.0223	0.7003	0.6993	0.6998	0.9252
2.3843	24000	0.0222	0.6971	0.7027	0.6999	0.9254
2.6823	27000	0.0219	0.7044	0.7004	0.7024	0.9259
2.9803	30000	0.0219	0.7047	0.7032	0.7040	0.9261

Framework Versions

Python: 3.10.8
SpanMarker: 1.5.0
Transformers: 4.28.0
PyTorch: 1.13.1+cu117
Datasets: 2.14.4
Tokenizers: 0.13.3

Citation

BibTeX

@software{Aarsen_SpanMarker,
    author = {Aarsen, Tom},
    license = {Apache-2.0},
    title = {{SpanMarker for Named Entity Recognition}},
    url = {https://github.com/tomaarsen/SpanMarkerNER}
}