bkmarzouk / ezpyp Goto Github PK

This could work in a similar sort of way to CI matrices, pytest parameterize, or some sort of "replicate" functionality...

In any case, we need an easier way to scale larger numbers of pipeline steps, which could be important, e.g. for sampling problems.

Currently execution prints - this would be better to log

Most of the components are in place to deal with this, but we essentially need to clean up any potentially corrupted files etc before relaunch