@<1523701295830011904:profile|CluelessFlamingo93> take a look here: None
the console_cr_flush_period
is actually quite internal and not often used, and it will only affect tasks running without an agent
And I'm looking at None as an example of a clearml.conf file and i can't seem to find sdk.development.worker.console_cr_flush_period
this flag.
Hi @<1523701295830011904:profile|CluelessFlamingo93> , if you self host then you're no longer limited on usage. However the downside is of course that you have to manage it yourself (Security, backups etc).
Hi @<1523701070390366208:profile|CostlyOstrich36> , I am using the community server, what happens if i change to a self hosting server?
I'm using Tensorboard to report everything, nothing special besides that.
I think these are the configurations you are looking for:
sdk.development.worker.report_event_flush_threshold
sdk.development.worker.console_cr_flush_period
inside clearml.conf
Wow, thanks a lot @<1523701070390366208:profile|CostlyOstrich36> for pointing me in the right direction. I also see that i can use sdk.development.worker.log_stdout
if i really need to kill my api calls before I'll Host my own server.
BTW what does suppress_update_message
do? I mean which kind of messages does it suppress?
I think there might be more configurations to control the amount of API calls you're making.
Hi @<1523701295830011904:profile|CluelessFlamingo93> , are you self hosting or using the community server?
I was actually referring to report_event_flush_threshold
@<1523701087100473344:profile|SuccessfulKoala55> in the file example here there is no reference to console_cr_flush_period