Automate and transfer chemical data extraction using span categorization and relation extraction models.
ChemREL is a PyPI (pip) package that allows you to train chemical data extraction models with ease using a suite of models, configurations, and data processing methods. ChemREL consists of a command line interface (CLI) through which you can run various commands, as well as a collection of different functions that you can import into your own code.
When utilizing or sharing the dataset and models included with ChemREL, kindly note that they are governed under the CC BY NC 4.0 license as imposed by software and data mining guidelines, permitting only non-commercial applications. They should be used exclusively for research.
Any alterations to the models, datasets, or functions included with ChemREL must be properly attributed using the citation provided in this documentation.
To view the full installation and usage reference for ChemREL, visit ChemREL's documentation.
The reference material on the documentation website can be additionally found as follows.
- ChemREL CLI Reference - Complete documentation of all available CLI commands.
- ChemREL Functions Reference - Complete documentation of ChemREL Python functions which are intended to be importable in your own code.
- Prediction Functions Demo Notebook - Jupyter notebook demonstrating usage of importable prediction functions.
- Span Categorization Demo Notebook - Jupyter notebook demonstrating the training of a new span categorization model.
Any alterations to the models, datasets, or functions included with ChemREL must be properly attributed according to the following citation.
@article{doi,
author = {Alshehri, Abdulelah S. and Horstmann, Kai A. and You, Fengqi},
title = {A Versatile Deep Learning Pipeline for Transferable Chemical Data Extraction},
journal = {Submitted to Journal of Chemical Information and Modeling},
volume = {n/a},
number = {n/a},
pages = {n/a},
keywords = {},
doi = {https://doi.org/},
url = {},
eprint = {},
abstract = {Abstract}
}