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! I'M Trying To Make Use Of Remote Execution In Clearml, But I Keep Getting An Error After The Agent Clones The Repo On The Remote Machine:

Hi everyone!
I'm trying to make use of remote execution in ClearML, but I keep getting an error after the agent clones the repo on the remote machine:

clearml_agent: ERROR: [Errno 36] File name too long: '/home/user/.clearml/venvs-builds/3.9/task_repository/repo.git/-m torch_ml.....fetch_and_train  --model-subtype ...'

It seems like the agent is trying to use part of the command I use to start the training as the file name, but this is failing as the command is long. Is there any way around this other than shortening the command? Happy to provide more info if needed, thanks!

  
  
Posted one year ago
Votes Newest

Answers 3


@<1523701070390366208:profile|CostlyOstrich36> Here is the task log and agent configuration

  
  
Posted one year ago

Hi John, this is for a new experiment but I'm running it by creating a new task in ClearML clearml.task.Task.init(...) and then calling task.execute_remotely(queue_name=...) . My goal is to run the same command I enter on a clearML agent instead of locally, but I can see in a lot of the documentation a pre-existing task is being cloned then queued - does remote execution work with new tasks as well? Give me a second to remove credentials etc from the logs/agent config and I'll send it over, thanks.

  
  
Posted one year ago

Hi @<1625666182751195136:profile|MysteriousParrot48> , how did you run the original experiment? Can you add the full log and also the agent configuration?

  
  
Posted one year ago
995 Views
3 Answers
one year ago
one year ago
Tags
Similar posts