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
Hi Everyone! Thanks To The Dev For The Excellent Work, I'M Migrating Some Work Projects To Clearml And It Has Been A Great Experience So Far. There Is Just One Thing I Cannot Understand From The Documentation: Can I Set Execution Details For A Pipeline E

Hi everyone! Thanks to the dev for the excellent work, I'm migrating some work projects to ClearML and it has been a great experience so far.

There is just one thing I cannot understand from the documentation: can I set execution details for a pipeline either from python API or from the "new run" button in UI?

Right now if I want to try my pipeline (defined in a python script) with e.g. a different python version, I have to create a draft from the python script, then in the UI modify the execution details for the draft and then run the draft. Is there a better way to do this?

  
  
Posted 7 months ago
Votes Newest

Answers 2


Did you find a solution?

  
  
Posted 7 months ago

Yes! I was looking at it wrong. The task execution details are only there for logging previous execution.
To modify the execution of the next run I had to modify the clearml agent. In this case I added this to the clearml.conf of my agent (I only have one worker right now):
python_binary: "/usr/local/bin/python3.10"
ignore_requested_python_version: true

  
  
Posted 7 months ago