BrightMosquito10 simply re-run it with the new version 🙂
What versions of ClearML/matplotlib are you using?
Sounds like some issue with queueing the experiment. Can you provide a log of the pipeline?
Can you provide a snippet to try and reproduce?
Hi RobustFlamingo1 ,
Can you point to where the website suggests that K8S is a requirement?
I use the ClearML-Agent on a local machine without any K8S. It is certainly not a requirement. From what I understand you can run it on K8S as well.
So to answer your question:
You can definitely use ClearML Orchestration (ClearML-Agent) with OR without K8S
I hope this helps 🙂
or /home/<USER_NAME>/clearml.conf
Try setting the following environment envs:%env CLEARML_WEB_HOST=
%env CLEARML_API_HOST=
%env CLEARML_FILES_HOST=
%env CLEARML_API_ACCESS_KEY=... %env CLEARML_API_SECRET_KEY=...
and try removing the clearml.conf file 🙂
Hi DilapidatedDucks58 , I think this might be a bug. Please open a GitHub issue to follow this 🙂
TenseOstrich47 , what do you mean exactly? Every task you run ends on 'aborted' status?
@<1523701235335565312:profile|HugeArcticwolf77> , I would suggest opening a feature request on GitHub for this 🙂
on /data/
Hi @<1537605927430000640:profile|NarrowSquirrel61> , how did you report the audio/image files?
Hi @<1570583227918192640:profile|FloppySwallow46> , can you please add the full log?
ScaryLeopard77 , Hi! Is there a specific reason to the aversion from pipelines? What is the use case?
"continue with this already created pipeline and add the currently run task to it"
I'm not sure I understand, can you please elaborate? (I'm pretty sure it's a pipelines feature)
Hi CrookedWalrus33 , do you have an ability to provide an absolute path?
I think if you provide an absolute path it should work 🙂
Hi @<1570220852421595136:profile|TeenyHedgehog42> , the docker images are stored together with all images - this is managed by docker and not clearml
Is the agent running on the same machine as the original code that didn't get any errors?
GiganticTurtle0 , just monitor the memory usage system wide
I think all relevant env variables are in the docker compose. You can spin different APIservers on different machines without elastic/mongo/redis and then point them to the relevant machines that have those services running
GiganticTurtle0 Hi!
Currently the usage shows the resources usage of the entire machine and not configurable otherwise. Regarding the refresh button - Currently it always auto refresh the graph every few seconds 🙂
I'll see if there is a way to make it report only the agent usage 🙂
Can you add a log?
Also, what happens if you apss it in agent.default_docker.arguments
?
Did you try what I added? Also the screenshot is too small, nothing is readable
Hi @<1526371965655322624:profile|NuttyCamel41> , try upgrading the Azure SDK package and try running again
Hi @<1585441140176326656:profile|StrongDove49> , what do you mean finding the credentials for each user? You mean on the databases?
Hi @<1545216070686609408:profile|EnthusiasticCow4> , I suggest you try ClearML-Serving
None