moleculenet icon indicating copy to clipboard operation
moleculenet copied to clipboard

Published 20 hours ago • verified publisher icon deepchem

Moleculenet data splits

Open ecvgit opened this issue 3 years ago • 3 comments

Is it possible to get the CSVs of Moleculenet data splits? I know it is possible to get it through the API, but for some reason dc.molnet.load_muv(splitter='random') takes a long time (few hours)! It would be nice to have this shared as a csv.

ecvgit avatar Aug 26 '21 18:08 ecvgit

It would definitely be useful to provide CSVs of splits! It's something we haven't gotten around to, but if anyone is interested in helping, please get in touch (the work will earn co-authorship on the upcoming MoleculeNet2 manuscript)

rbharath avatar Sep 08 '21 16:09 rbharath

I generated the CSVs couple of days back. Would be glad to share it. Should I create a PR adding the CSVs to here? https://github.com/deepchem/deepchem/tree/master/datasets

ecvgit avatar Sep 08 '21 20:09 ecvgit

@ecvgit Great! Could you contribute it to this repository? This is probably the correct home for the time being

rbharath avatar Sep 08 '21 22:09 rbharath