Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Answered
Hello, I'M Confused About The Best Way To Use A Docker Container To Manage The Environment For Experiments. I'Ve Gone Through Some Of The Tutorials Running In Venv Mode And Am Now Trying To Run Some Of My Own Codebase. I Have A Docker Container With All T

Hello, I'm confused about the best way to use a docker container to manage the environment for experiments. I've gone through some of the tutorials running in venv mode and am now trying to run some of my own codebase. I have a docker container with all the dependencies. Should I just specify my clearml instrumented script as the entrypoint to the container, or is there some other way?

  
  
Posted 9 months ago
Votes Newest

Answers 3


then will clearml associate that image with my experiment and always use that image with it,

when you say "agent to use my docker image," I'm assuming you mean the configuration file or --docker argument, in both cases this means Default conatiner.
This means that if the Task does Not specify a docker, it will use the one you set in the conf/argument, But Tasks can always specify a different docker to use, and the agent will pull the requested docker based on the Task's entry.
Even when you clone a Task you have the previously used docker image set, but you can still change this value in the UI.

  
  
Posted 9 months ago

Hi @<1653207659978952704:profile|LovelyStork78>

I have a docker container with all the dependencies.

Well I think the main question is are you using the clearml-agent to launch jobs/experiments? If you do it makes sense to specify your docker as "base docker image" (in the UI look for under the Execution tab, Container).
This means the agent will use the pre-installed environment and will add anything that your Task needs on top of it, this of course includes pushing your codebase in runtime into the container, so that you do not need to have a container per job. Does that make sense ?

  
  
Posted 9 months ago

Hello, @<1523701205467926528:profile|AgitatedDove14> . Thank you for the reply. Yes, I am using an agent (and a self-hosted server). So, if I set up my agent to use my docker image, then will clearml associate that image with my experiment and always use that image with it, say if I cloned it?

  
  
Posted 9 months ago
572 Views
3 Answers
9 months ago
9 months ago
Tags