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, Is There A Way To Check Whether The Experiment Was Run From Cli (E.G. Using Clearml-Task) Or By Clone+Enqueue From The Api, I Mean From The Code While The Experiment Is Running?

Hello, is there a way to check whether the experiment was run from CLI (e.g. using clearml-task) or by clone+enqueue from the API, I mean from the code while the experiment is running?

  
  
Posted one year ago
Votes Newest

Answers 6


ResponsiveHedgehong88 hi,

The best indication would be in the 'INFO' section of the experiment. If it was run via CLI it should have N/A in the worker/queue section

  
  
Posted one year ago

ResponsiveHedgehong88 , please look here:
https://clearml.slack.com/archives/CTK20V944/p1660142477652039
Is this what you're looking for?

  
  
Posted one year ago

ResponsiveHedgehong88 , what about using tags? You can tag experiments and this way you can easily tell them apart

  
  
Posted one year ago

Ok thanks 🙂

  
  
Posted one year ago

CostlyOstrich36 hi, thanks for the answer. Unfortunately, both my CLI run experiment (using clearml-task ) and the one cloned from this have the same setup in the INFO/worker and INFO/queue. I do set the queue as one of the clearml-task arguments.

  
  
Posted one year ago

CostlyOstrich36 no, the code is always running remotely. I use two ways to start an experiment. The first is using clearml-task interface https://github.com/allegroai/clearml/blob/master/docs/clearml-task.md where I define which script to run from a repo + branch, and also define a queue (remote worker) for where to run the experiment. The second way is by manually clicking Clone in the API dashboard (then modifying some params in configuration objects) then setting Enqueue to the same remote worker.

  
  
Posted one year ago