jmansour
commented
5 years ago Yeah certainly more than happy for it to be generalised. I'm the case of the line above, as long as it writes out a unique job I'd, that's fine. The Id is necessary as it allows us to run identical jobs numerous times so that we can average over the results.
On Fri, 12 Jul. 2019, 12:16 pm Julian Giordani, notifications@github.com wrote:
I'm just wondering how critical this is? I was thinking of removing all raijin / pawsey / environment specific code and putting it in the script that invokes qsub or sbatch. Does that seem right to you @jmansour https://github.com/jmansour ? I'm happy to do it. Just wanted to run it by you.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/jmansour/scaling_scripts/issues/1?email_source=notifications&email_token=AAK7NHMNCGM7PUIQOZUCZBTP67SPVA5CNFSM4IB6VTBKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4G6YWQUQ, or mute the thread https://github.com/notifications/unsubscribe-auth/AAK7NHI6SJDTE2ZM5SPXF43P67SPVANCNFSM4IB6VTBA .
https://github.com/jmansour/scaling_scripts/blob/dcdb83c9897056d0193e1d6e5fb0d92958c18c7a/rt_timed.py#L213
I'm just wondering how critical this is? I was thinking of removing all raijin / pawsey / environment specific code and putting it in the script that invokes qsub or sbatch. Does that seem right to you @jmansour ? I'm happy to do it. Just wanted to run it by you.