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
Unanswered
Hi, I Try To Execute Pipeline With Pipelinecontroller And Define It Like This: Pipe = Pipelinecontroller(


yes thanks , but if I do this, the packages will be installed for each step again, is it possible to use a single venv?

Notice that the venv is Cached on the clearml-agent host machine (if this is k8s glue, make sure to setup the Cache as a PV to achieve the same)
This means there is no need to worry about that and this is stable.
That said, if you have an existing VENV inside the container, just add docker_args="-e CLEARML_AGENT_SKIP_PIP_VENV_INSTALL =/path/to/bin/python"
See here: None

BTW: notice that in your example you are probably installing to the system path not the venv path, and to your point specifically, add ENV=CLEARML_AGENT_SKIP_PIP_VENV_INSTALL=$VIRTUAL_ENV

Lastly if you want your agent to Alwoays use the activated Venv inside the container set:

docker_use_activated_venv=true

None

  
  
Posted one month ago
32 Views
0 Answers
one month ago
one month ago