Skip to main content

oganesson enables rapid AI workflows for material science and chemistry

Project description

Oganesson

oganesson (og for short) is a python package that enables you to apply artificial intelligence workflows to your material discovery projects.

Installation

og requires the installation of the following library:

After installing the above library, you can install og using the pip command as follows:

pip install oganesson

Features

og is currently under active development. The following features are currently available.

Machine learning descriptors

og will bring together machine learning descriptors for materials and molecules within a unified framework. og currently provides the following descriptors:

  • The BACD, ROSA and SymmetryFunctions introduced in this publication
  • Most of the descriptors from DScribe

Each descriptor has its own class, which extends the Descriptors class in the oganesson.descriptors module. Here is an example of how to describe a structure using the BACD and SymmetryFunctions descriptor classes.

from oganesson.descriptors import BACD, SymmetryFunctions
from oganesson.ogstructure import OgStructure

bacd = BACD(OgStructure(file_name='examples/structures/Li3PO4_mp-13725.cif'))
print(bacd.describe())

sf = SymmetryFunctions(OgStructure(file_name='examples/structures/Li3PO4_mp-13725.cif'))
print(sf.describe())

Genetic algorithms

The main purpose of og is to make complex artificial intelligence workflows easy to compose. Here is an example: running a genetic search for structures, where the structure optimization is performed using the M3GNET neural network model.

from oganesson.genetic_algorithms import GA
ga = GA(species=['Na']*4 + ['H']*4)
for i in range(10):
    ga.evolve()

Generation of the diffusion path for NEB calculations

The most painful part of doing transition state calculations in VASP is in building the images. The following code makes this happen in 2 lines of code. You only need to specify the structure file, and the atomic species you want to diffuse, and OgStructure will generate a folder for each path, and then write the POSCAR image files in each of these folders.

In the following example, we explore the possible Li diffusion paths in Li3PO4, given there are 6 Li atoms in the cell.

from oganesson.ogstructure import OgStructure
og = OgStructure(file_name='examples/structures/Li3PO4_mp-13725.cif')
og.generate_neb('Li')

Note that the default value of r, which is 3, is sufficient for lithium systems. However, for the case of larger atoms such as Na, a larger value of r would be required.

Finding a reasonable adsorption site of an atom on a surface

from oganesson.ogstructure import OgStructure
og=OgStructure(file_name='examples/structures/MoS2.vasp')
og.add_atom_to_surface('Li').structure.to('MoS2_Li.vasp','poscar')

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

oganesson-0.1.43.tar.gz (1.3 MB view details)

Uploaded Source

Built Distribution

oganesson-0.1.43-py3-none-any.whl (1.3 MB view details)

Uploaded Python 3

File details

Details for the file oganesson-0.1.43.tar.gz.

File metadata

  • Download URL: oganesson-0.1.43.tar.gz
  • Upload date:
  • Size: 1.3 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.0.0 CPython/3.11.9

File hashes

Hashes for oganesson-0.1.43.tar.gz
Algorithm Hash digest
SHA256 3489a667969f5052c73a6a266aa24b6ae0b399c26406ba6ae3dd2e5f9c422046
MD5 0e53bb575dd66cac5d9e3bf244e4e5f5
BLAKE2b-256 da8b1d4889ea496638590f1b58cdfaf5f702298cc0ef7a72c5e76e44d2ad3e12

See more details on using hashes here.

File details

Details for the file oganesson-0.1.43-py3-none-any.whl.

File metadata

  • Download URL: oganesson-0.1.43-py3-none-any.whl
  • Upload date:
  • Size: 1.3 MB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.0.0 CPython/3.11.9

File hashes

Hashes for oganesson-0.1.43-py3-none-any.whl
Algorithm Hash digest
SHA256 8e775410ce1935982bcd707786eed11af6d4da7d0ac5661422717898ff1c9eff
MD5 9cb1af78b7809a29cdc72d1f328ca8a1
BLAKE2b-256 99074a2e14bdbbdff3ed71394c1496f94263dce327c1a5a2cc0a6cd5fa25279e

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page