Fig. 1

Pathway used to assemble and distribute the GriddingMachine database. Each dataset is (a) reprocessed to meet our standards for distribution, (b) compressed as a tar.gz file along with an empty label file GRIDDINGMACHINE, and (c) stored on publicly available HTTP servers. Then the meta information for each dataset is stored in Artifacts.toml, which includes the tag name, sha1 hashtag, sha256 hashtag, and downloading URLs. Users are able to access the datasets manually via HTTP protocols or automatically through Julia, Matlab, Octave, Python, and R functions aided by Artifacts.toml.