PyQAlloy-compatible Model for RMSAD prediction based on [Tandoc 2023 (10.1038/s41524-023-00993-x)](https://doi.org/10.1038/s41524-023-00993-x)
Project description
Lattice-Distortion
This Fork
This small repository is a lightweight fork version of the original one by Tandoc. It is only slightly modified to fit automated pipelines of the ULTERA Database (ultera.org) infrastrcutre, which expects model.py
script with predict(comp: pymatgen.core.Composition)
function returning an ordered array of output numbers or a labeled dictionary of them.
Original README by Tandoc
This repository contains relevant code and data for "Mining lattice distortion, strength, and intrinsic ductility of refractory high-entropy alloys using physics-informed >statistical learning" by Christopher Tandoc, Yong-Jie Hu, Liang Qi, and Peter K. Liaw to be published in npj Computational Materials
RMSAD_tool.py is a linux command line script written in python that takes a chemical composition in the form of a text string and prints the lattice distortion in angstroms.
example usgage: ./RMSAD_tool.py Ti0.5V0.5
-This script uses pymatgen (https://pymatgen.org/) to process the input string and is thus a requirement for the script to work. Depending on the version of pymatgen you have >installed, lines 3 and 380 may need to be modified (https://matsci.org/t/python-problem-with-pymatgen/35720) -numpy (https://numpy.org/) is also a dependency -This tool is currently only able to make predictions for compositions containing Ti,Zr,Hf,V,Nb,Ta,Mo,W,Re,Ru and will return an error if any other elements are present in >the input -B2 and elemental feature data are defined in dictionaries at the beginning of the code
training.ipynb and training_data.csv contains code and data to reproduce the rmsad model training that was performed in the paper -jupyter notebook is needed to open training.ipynb, dependencies are numpy, pymatgen, matplotlib (https://matplotlib.org/), pandas (https://pandas.pydata.org/), and >sklearn(https://scikit-learn.org/stable/)
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for pqam_rmsadtandoc2023-0.1.0.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 0f6cba3488d7e908605721b4b15815dad10001198d17ff6160a9ee3b0c23d769 |
|
MD5 | 1178b8c03e8177378622550aabbac420 |
|
BLAKE2b-256 | 024f7981267252a0613a7dd6b63d8590c65840766a22adf10a48de3acdd82f5a |
Hashes for pqam_rmsadtandoc2023-0.1.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | bbf46f9d8c86a30b1ba2b070106d7b2b128591b1998f45155bbe915671889e45 |
|
MD5 | 89cc52a1f340cd2a8ea9006beee60939 |
|
BLAKE2b-256 | c6480fb36d4717858b02c035f27937268cabf8767b997be90788af511b990a58 |