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, Does Anyone Know How To Bypass Package Management By Clearml If Using Docker Mode? I Want To Achieve The Following -

Hello, Does anyone know how to bypass package management by clearml if using docker mode? I want to achieve the following -
Run agent in docker mode Push a task in queue on which the above agent is listening A container is deployed and no venvs are created. I want to then run my own scripts for package installation and not use a requirements file or provide --package (python packages). The task gets executedHas anyone tried to use their own package installation scripts instead of relying on clearml's package management system?

  
  
Posted 2 years ago
Votes Newest

Answers 7


Ideally I want to use conda environment.yml instead of a requirements file.

  
  
Posted 2 years ago

Amazing. Thanks! Is there a similar setting available for conda mode.

  
  
Posted 2 years ago

AgitatedDove14 You are right. I confused myself by making a minor error in passing flags.

  
  
Posted 2 years ago

you can do it with docker_bash_setup_script where you run conda install what you need

  
  
Posted 2 years ago

but how pass the file as argument, idk

  
  
Posted 2 years ago

Pass this to the func_step
docker_args="--env CLEARML_AGENT_SKIP_PYTHON_ENV_INSTALL=1",

  
  
Posted 2 years ago

TroubledHedgehog16 if you have a preinstalled conda env then why would you need to reinstall it from yml file? Also if this is the default python env, clearml-agent will inherit from it and use i, (no real overhead there)
Notice the reason for "inheriting system" python environments is so that the agent could cache the individual Task requirements, meaning next time it will not need to reinstall anything
wdyt?

  
  
Posted 2 years ago
1K Views
7 Answers
2 years ago
one year ago
Tags