SpanMarker with numind/generic-entity_recognition-v1 on FewNERD

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 Sources

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

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
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}
}
Downloads last month
302
Safetensors
Model size
125M params
Tensor type
I64
·
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for guishe/span-marker-generic-ner-v1-fewnerd-fine-super

Base model

numind/NuNER-v0.1
Finetuned
(4)
this model

Dataset used to train guishe/span-marker-generic-ner-v1-fewnerd-fine-super

Collection including guishe/span-marker-generic-ner-v1-fewnerd-fine-super

Evaluation results