This repository contains machine learning models of Named Entity Recognition (NER), designed to be deployed using ONNX and utilized in a Streamlit-based web application. The app provides an interactive interface for performing this task using neural network architectures. Check here to see other ML tasks.
For more information about the training process, please check the ner.ipynb file in the training folder.
If you encounter message This app has gone to sleep due to inactivity, click Yes, get this app back up! button to wake the app back up.
If the demo page is not working, you can fork or clone this repository and run the application locally by following these steps:
-
Clone the repository:
git clone https://github.com/verneylmavt/st-ner.git cd st-ner -
Install the required dependencies:
pip install -r requirements.txt
-
Run the Streamlit app:
streamlit run app.py
Alternatively you can run jupyter notebook demo.ipynb for a minimal interface to quickly test the model (implemented w/ ipywidgets).
I acknowledge the use of the CoNLL-2003 dataset provided by the Conference on Computational Natural Language Learning (CoNLL). This dataset has been instrumental in conducting the research and developing this project.
- Dataset Name: CoNLL-2003
- Source: https://www.aclweb.org/anthology/W03-0419/
- Description: This dataset was introduced as part of the CoNLL-2003 shared task on language-independent named entity recognition. It includes annotated data for four types of named entities: persons (PER), locations (LOC), organizations (ORG), and miscellaneous names (MISC). The dataset covers English and German languages and is widely used for training and evaluating NER systems.
I deeply appreciate the efforts of the CoNLL organization in making this dataset available.