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
[Instance Autoscaler For Gcp]

[Instance AutoScaler for GCP]
In case someone else is interested, we have build an AutoScaler for GCP, too.
It works similar to the AWS one in the ClearML repo by creating a GCPDriver (similar to the AWS Driver) and starting it in a script.

Note that in this case, we stayed away from sending and variables, scripts or installs to the instance, which makes the GCPDriver much slimmer.
Instead, we create a suitable machine image “by hand” in GCP that is set up to start the ClearML Agent correctly when started, and the AutoScaler just starts and stops instances with this image.

  
  
Posted one year ago
Votes Newest

Answers 3


Hi @<1523702496097210368:profile|ScantChimpanzee51> , would you be interested in adding a PR to the agent with this in the examples?

  
  
Posted one year ago

Hi Jake, yes I’d love to! Just a question: how clean and complete does the example need to be? For example, this code relies on you building a correct Machine Image on GCP (which is somewhat unrelated to ClearML) and it does not get the logs from the agent instances - is that still good enough?

  
  
Posted one year ago

Seems good enough for me, as long as all that is well documented 🙂

  
  
Posted one year ago