license: apache-2.0
base_model: distilbert-base-cased
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: distilbert-NER
results: []
distilbert-NER
Model description
distilbert-NER is the fine-tuned version of DistilBERT, which is a distilled variant of the BERT model. DistilBERT has fewer parameters than BERT, making it smaller, faster, and more efficient. distilbert-NER is specifically fine-tuned for the task of Named Entity Recognition (NER).
This model accurately identifies the same four types of entities as its BERT counterparts: location (LOC), organizations (ORG), person (PER), and Miscellaneous (MISC). Although it is a more compact model, distilbert-NER demonstrates a robust performance in NER tasks, balancing between size, speed, and accuracy.
The model was fine-tuned on the English version of the CoNLL-2003 Named Entity Recognition dataset, which is widely recognized for its comprehensive and diverse range of entity types.
Available NER models
| Model Name | Description | Parameters |
|---|---|---|
| bert-base-NER | Fine-tuned BERT-base model for NER - balanced performance | 110M |
| distilbert-NER | Fine-tuned DistilBERT - smaller, faster, lighter than BERT | 66M |
Intended uses & limitations
How to use
This model can be utilized with the Transformers pipeline for NER, similar to the BERT models.
from transformers import AutoTokenizer, AutoModelForTokenClassification
from transformers import pipeline
tokenizer = AutoTokenizer.from_pretrained("dslim/distilbert-NER")
model = AutoModelForTokenClassification.from_pretrained("dslim/distilbert-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)
Limitations and bias
The performance of distilbert-NER is linked to its training on the CoNLL-2003 dataset. Therefore, it might show limited effectiveness on text data that significantly differs from this training set. Users should be aware of potential biases inherent in the training data and the possibility of entity misclassification in complex sentences.
Training data
The model was fine-tuned on the English version of the standard CoNLL-2003 Named Entity Recognition dataset, known for its effectiveness in training NER models.
Training procedure
The training details, including hardware specifications, aren't specified. However, the model's training followed the best practices suitable for distilbert models, aiming at an efficient balance between learning efficiency and model accuracy.
Eval results
| Metric | Score |
|---|---|
| Loss | 0.0710 |
| Precision | 0.9202 |
| Recall | 0.9232 |
| F1 | 0.9217 |
| Accuracy | 0.9810 |
The training and validation losses demonstrate a decrease over epochs, signaling effective learning. The precision, recall, and F1 scores are competitive, showcasing the model's robustness in NER tasks.
BibTeX entry and citation info
For DistilBERT:
@article{sanh2019distilbert,
title={DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter},
author={Sanh, Victor and Debut, Lysandre and Chaumond, Julien and Wolf, Thomas},
journal={arXiv preprint arXiv:1910.01108},
year={2019}
}
For the underlying BERT model:
@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 = {db
lp computer science bibliography, https://dblp.org}
}