If my open source models have been useful to you, please consider supporting me in building small, useful AI models for everyone (and help me afford med school / help out my parents financially). Thanks!
bert-base-NER is a fine-tuned BERT model that is ready to use for Named Entity Recognition and achieves state-of-the-art performance for the NER task. It has been trained to recognize four types of entities: location (LOC), organizations (ORG), person (PER) and Miscellaneous (MISC).
Specifically, this model is a bert-base-cased model that was fine-tuned on the English version of the standard CoNLL-2003 Named Entity Recognition dataset.
If you'd like to use a larger BERT-large model fine-tuned on the same dataset, a bert-large-NER version is also available.
You can use this model with Transformers pipeline for NER.
from transformers import AutoTokenizer, AutoModelForTokenClassification
from transformers import pipeline
tokenizer = AutoTokenizer.from_pretrained("dslim/bert-base-NER")
model = AutoModelForTokenClassification.from_pretrained("dslim/bert-base-NER")
nlp = pipeline("ner", model=model, tokenizer=tokenizer)
example = "My name is Wolfgang and I live in Berlin"
ner_results = nlp(example)
print(ner_results)
This model is limited by its training dataset of entity-annotated news articles from a specific span of time. This may not generalize well for all use cases in different domains. Furthermore, the model occassionally tags subword tokens as entities and post-processing of results may be necessary to handle those cases.
The training dataset distinguishes between the beginning and continuation of an entity so that if there are back-to-back entities of the same type, the model can output where the second entity begins. As in the dataset, each token will be classified as one of the following classes:
Abbreviation|Description -|- O|Outside of a named entity B-MISC |Beginning of a miscellaneous entity right after another miscellaneous entity I-MISC | Miscellaneous entity B-PER |Beginning of a person’s name right after another person’s name I-PER |Person’s name B-ORG |Beginning of an organization right after another organization I-ORG |organization B-LOC |Beginning of a location right after another location I-LOC |Location
This dataset was derived from the Reuters corpus which consists of Reuters news stories. You can read more about how this dataset was created in the CoNLL-2003 paper.
This model was trained on a single NVIDIA V100 GPU with recommended hyperparameters from the original BERT paper which trained & evaluated the model on CoNLL-2003 NER task.
The test metrics are a little lower than the official Google BERT results which encoded document context & experimented with CRF. More on replicating the original results here.
@article{DBLP:journals/corr/abs-1810-04805,
author = {Jacob Devlin and
Ming{-}Wei Chang and
Kenton Lee and
Kristina Toutanova},
title = {{BERT:} Pre-training of Deep Bidirectional Transformers for Language
Understanding},
journal = {CoRR},
volume = {abs/1810.04805},
year = {2018},
url = {http://arxiv.org/abs/1810.04805},
archivePrefix = {arXiv},
eprint = {1810.04805},
timestamp = {Tue, 30 Oct 2018 20:39:56 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-1810-04805.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
@inproceedings{tjong-kim-sang-de-meulder-2003-introduction,
title = "Introduction to the {C}o{NLL}-2003 Shared Task: Language-Independent Named Entity Recognition",
author = "Tjong Kim Sang, Erik F. and
De Meulder, Fien",
booktitle = "Proceedings of the Seventh Conference on Natural Language Learning at {HLT}-{NAACL} 2003",
year = "2003",
url = "https://www.aclweb.org/anthology/W03-0419",
pages = "142--147",
}
Files
18 files · 2.16 GBWeights 433 MB (safetensors). Sizes and sha256 from the signed manifest.
