A document-level information extraction pipeline for layered cathode materials for sodium-ion batteries.
Project description
CathodeDataExtractor
Cathodedataextractor
is a lightweight document-level information extraction pipeline that can automatically extract
comprehensive properties related to synthesis parameters, cycling and rate performance of cathode materials from the
literature of layered cathode materials for sodium-ion batteries.
Installation
pip install cathodedataextractor
Features
- It is built on open-source libraries: pymatgen, text2chem, and ChemDataExtractor v2 with some modifications.
- BatterySciBERT-uncased Multi-Label text classification model for filtering documents.
- Automated comprehensive data extraction pipeline for cathode materials.
- Paragraph Multi-Class classification algorithms for documents (HTML/XML) from the RSC and Elsevier.
- A normalised entity handling process is provided.
- An effective chemical abbreviation detection module.
- Heuristic multi-level relation extraction algorithm for electrochemical properties.
In addition, the pipeline is also suitable for string sequence text extraction.
Quick start
Extract from documents
from glob import iglob
from cathodedataextractor.information_extraction_pipe import Pipeline
pipline = Pipeline()
for document in iglob('*ml'):
extraction_results = pipline.extract(document)
Extract from string
from cathodedataextractor.information_extraction_pipe import Pipeline
extraction_results = Pipeline.from_string(
'Apart from the conventional cationic redox of transition metals, '
'both Na-deficit and Na-excess materials have showcased the ability '
'to exploit oxygen redox activity as O2–/O2n– for a charge '
'compensation mechanism. To realize cathodes with enhanced energy '
'density, a technique like the incorporation of alkali metal ions '
'into transition metal layers has been adopted. Recent work by Boisse '
'(13) et al. displayed the impact of honeycomb cation ordering of '
'a highly stabilized intermediate phase for a Na2RuO3 cathode material '
'in instigating the anionic redox activity and providing a capacity '
'of 180 mAh g–1 at 0.2C with a capacity retention of 89% for over '
'50 cycles. More devoted efforts to realize the utmost potential '
'of anionic redox ought to be carried out in the future.')
Issues?
You can either report an issue on GitHub or contact me directly. Try gouyx@mail2.sysu.edu.cn.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
cathodedataextractor-0.0.3.tar.gz
(44.1 kB
view hashes)
Close
Hashes for cathodedataextractor-0.0.3.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 29e17300f5ca49f80a8e13d59d3dab78044c0973fe55d6d82e633996f44470fb |
|
MD5 | 29e37f23bd439612301f0a588ee5dee8 |
|
BLAKE2b-256 | ad04cc5fb965cee54c2afd10c5c86ec72f963e33b14d56165284f8ecd6c08030 |