Skip to content

materialscloud-org/archive-data-miner

Repository files navigation

Materials Cloud Structure data mining.

Structured data mining of the materials cloud archive.

Currently we show 4 stats on the main page:

  • Number crystal structures
  • structures with assoiciated DFT SCF calcualtions (seems unclear)
  • data and calculation nodes in the explore section
  • reproducible calculations in the explore section

How to mine

This repository is designed to not take 500 hours on recalculation and thus the scripts reads from the cache files i.e RESULTS-CACHE.YAML...

NOTE: THE ARCHIVE CURRENTLY RELIES ON A WIP FEATURE IN AIIDA - #PR7433. The installed aiida-core is from this commit: #372ae45 ...

Explorer sections

To recalculate the Explore section parts you need to update the EXPLORE-LINKS.YAML with new explore entries (the structure matches that seen in mc-frontend for simplicity.)

entries:
    - title: mc3d-pbe-v1:
      doi:
        - id: DOI ID (partial or http link to the archive)
          type: !! UNUSED FIELD.
      description: !! UNUSUED FIELD

Once done updating the yaml run the script ./mine-explore-sections.py, this will check if the results exist in EXPLORE-RESULTS-CACHE.YAML, if so it will skip them, otherwise it will pull the archive (potentially very big) and then perform an aiida query to determine the number of structures.

About

For mining Archive .aiida entries.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors