Hi @<1686547380465307648:profile|StrongSeaturtle89> , yes 🙂
@<1718799873618219008:profile|FunnyPeacock68> , it appears that reading a yaml isn't supported. Currently only requirements.txt. I'd suggest opening a GitHub feature request for this capability!
CluelessElephant89 , Hi 🙂
An experiment cannot be 'unpublished' through the UI since it's a 'finalizing' action that the UI also gives a popup regarding. I think you can reset the task with Task.reset(force=True) through the SDK. This however will remove all logs/metrics.
https://clear.ml/docs/latest/docs/references/sdk/task#reset
What is the reason you would like to 'un-publish' and what would you want the new state to be, completed?
Yes. However I don't think you need to backup Redis as it holds only data related to the currently running agents.
You will need to set agent.git_user agent.git_pass in your clearml.conf so the agent will be able to clone the repository. For git_pass you will need to use your github token or bitbucket app password
Hi @<1539417873305309184:profile|DangerousMole43> , if understand correctly, you basically want to have some logic in your pipeline that will launch jobs according to some "trigger"?
CluelessElephant89 , well that looks like a bug
I'll sort it out on our end 🙂
You could try using the API reference:
https://clear.ml/docs/latest/docs/references/api/tasks#post-tasksreset
Already tested that it works. However please notice that it will reset the task removing ALL outputs
Can you try with the latest version of the server?
Hi @<1533620191232004096:profile|NuttyLobster9> , are you self hosting ClearML?
Do you see any errors in the mongo or elastic containers when you create a dataset?
Also, does this issue reproduce when you create datasets via the CLI rather then the SDK?
Hi @<1730396272990359552:profile|CluelessMouse37> , I would suggest reviewing the relevant docs regarding pipelines & HPO and then running said examples before integrating them 🙂
None
None
None
[None](https://clear.ml/docs/latest/docs/references/sdk/hpo_optimizati...
@<1590514584836378624:profile|AmiableSeaturtle81> , its best to open a GitHub issue in that case to follow up on this 🙂
Hi @<1739093605621960704:profile|LovelySparrow29> , there is no such schema currently. Out of curiosity, what is the use case?
@<1739093605621960704:profile|LovelySparrow29> , forgot to add the docs as well - None
You can add torch to the installed packages section manually to get it running but I'm curious why it wasn't logged. How did you create the original experiment?
Please check the task__trash collection in mongodb backend database. If you find all your tasks there then someone indeed deleted them
Hi @<1578555761724755968:profile|GrievingKoala83> , from my understanding this is a feature to be added to the imminent release of clearml-serving
Hi RotundHedgehog76 , from API perspective I think you are correct
How did you setup the ClearML server?
What do you mean by public to private mongo? @<1734020208089108480:profile|WickedHare16>
Hi @<1691983266761936896:profile|AstonishingOx62> , what if you try directly with the REST api and no the APIClient?
Hi HarebrainedBaldeagle11 , I think OSS is bring your own security so it's probably possible
Hi @<1554638179657584640:profile|GlamorousParrot83> , yes it is supported just like in the AWS autoscaler 🙂
Hi IdealGorilla64 , can you please provide a screenshot? How much smaller than 200 have you tried. And finally which version of ClearML are you using? 🙂
Hi DullPeacock33 , I think what you're looking for is this:
https://clear.ml/docs/latest/docs/references/sdk/task#execute_remotely
This will initialize all the automagical stuff but won't require running the script locally.
What do you think?
What version of clearml-agent are you using? Can you add the full log here?