o-mfml_for_qc

Code and data to accompany mnauscript titled "Optimized Multi-Fidelity Machien Learning for Quantum Chemistry"

https://github.com/sm4da/o-mfml_for_qc

Science Score: 49.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
    Found 2 DOI reference(s) in README
  • Academic publication links
    Links to: arxiv.org, acs.org
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (6.1%) to scientific vocabulary
Last synced: 8 months ago · JSON representation

Repository

Code and data to accompany mnauscript titled "Optimized Multi-Fidelity Machien Learning for Quantum Chemistry"

Basic Info
  • Host: GitHub
  • Owner: SM4DA
  • License: mit
  • Language: Jupyter Notebook
  • Default Branch: master
  • Size: 113 MB
Statistics
  • Stars: 1
  • Watchers: 2
  • Forks: 2
  • Open Issues: 0
  • Releases: 0
Created over 2 years ago · Last pushed about 1 year ago
Metadata Files
Readme License Citation

README.md

Optimized Multi-Fidelity Machine Learning for Quantum Chemistry

This repository contains the scripts and data to reproduce the results of the work by Vinod et. al. titled "Optimized Multi-Fidelity Machine Learning for Quantum Chemistry" (available at [https://arxiv.org/abs/2312.05661]). The raw data of molecules for the QM7b dataset can be downloaded from [https://achs-prod.acs.org/doi/10.1021/acs.jctc.8b00832#articlecontent-right]. The rawdata for the Excitation State Energies can be downloaded from [https://github.com/SM4DA/MultiFidelityMachineLearning-for-MolecularExcitationEnergies] with explanation present in Vinod _et. al. (2023) available at [https://pubs.acs.org/doi/10.1021/acs.jctc.3c00882].

The scripts in this repository and the plots they reproduce are listed below: * QM7b/GenerateSLATM.py generates the Global SLATM representation for the 7211 molecules of the QM7b data. * QM7b/LearningCurves_QM7b.py generates data to reproduce Figure 3-5 of the main manuscript and Figure 1 of the supplementary text. * QM7b/pople_MFML_outs.py generates the single fidelity learning curve from these figures. * QM7b/Coeff_analysis_removed_fidelity.py compares the full o-MFML model and reduced o-MFML model as per the analysis of hte coefficients. * ExcitedState/LearningCurves_ExcitedState.py generates data for Figure 6,7 of the main text, and Figure 2,3 of the Supplementary text. * ExcitedState/CompareMFMLtypes.py generates data for Table 1 in the supplementary text.

All the plotting routines for the QM7b segment are found in QM7b/QM7bPlots.ipynb and those for the Excitation state can be found in ExcitedState/ExcitedStatePlots.ipynb.

Owner

  • Name: Software for Data-Intensive Applications Group @ University of Wuppertal
  • Login: SM4DA
  • Kind: organization

GitHub Events

Total
Last Year

Dependencies

requirements.txt pypi
  • conda =4.12.0
  • jupyterlab =3.3.2
  • numpy =1.24.3
  • pip =23.2.1
  • python =3.9.18
  • qml =0.4.0.27
  • sns =0.12.2
  • tqdm =4.65.0