You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Since the session class is a pretty good do-it-all script, it wouldn't be too difficult to add a command-line interface (CLI) to PBjam - useful when running on remote workstations or via job schedulers. I'd be happy to have a crack at this when I have time, converting some things from my repo https://github.com/alexlyttle/jam-ensemble (with the recent help from @nielsenmb). I have recently done something like this, albeit lightweight, on a personal repo. However, I've posted this issue here in case anyone else is more experienced to write something like this.
I'd imagine this is not high priority at the minute, just a suggestion.
where --njobs could default to the length of the input csv if not specified (i.e. one job per star). Could have a --template option for custom script templates and an option to write scripts to individual files in the output directory as opposed to pbjam-session.tot.
Once we have a module which does this (e.g. named pipeline.py) it can be added to setup.py as the following argument, e.g.
Since the
session
class is a pretty good do-it-all script, it wouldn't be too difficult to add a command-line interface (CLI) to PBjam - useful when running on remote workstations or via job schedulers. I'd be happy to have a crack at this when I have time, converting some things from my repo https://github.com/alexlyttle/jam-ensemble (with the recent help from @nielsenmb). I have recently done something like this, albeit lightweight, on a personal repo. However, I've posted this issue here in case anyone else is more experienced to write something like this.I'd imagine this is not high priority at the minute, just a suggestion.
It would work something like this (similar to https://github.com/danxhuber/isoclassify CLI) for an individual star, e.g
for a single jam session on all the stars in inputs.csv, e.g.
or generate batch scripts for multiple sessions, e.g.
$ pbjam batch --njobs=100 -i /path/to/inputs.csv -o /path/to/output > pbjam-session.tot
where
--njobs
could default to the length of the input csv if not specified (i.e. one job per star). Could have a--template
option for custom script templates and an option to write scripts to individual files in the output directory as opposed to pbjam-session.tot.Once we have a module which does this (e.g. named pipeline.py) it can be added to
setup.py
as the following argument, e.g.allowing it to be added to the user's path upon install (something I have done in https://github.com/alexlyttle/isoclassify).
The text was updated successfully, but these errors were encountered: