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, I Was Wondering If There Is A Way To Attach Files That Are Not Committed To Git To Task Run? I'M Aware To The Option Of

Hi, i was wondering if there is a way to attach files that are not committed to git to task run? i'm aware to the option of apply_files in the clearml.conf file, which demands writing the content of the file in the conf file. i'm looking for a way to easily work with uncommitted configuration files, why are they not committed? in out repo setup we have a template configuration files, which each user manipulates, when introducing changes to these templates, merging the local config file with the updates from git is not so easy. i would like to decouple the template config and the actual used configs (which are not uploaded to git) to gain better UX. is it supported in clearml?

  
  
Posted 4 months ago
Votes Newest

Answers 4


@<1822805241150574592:profile|ShinySparrow39> thanks for replying but i think you got me wrong. in your suggestion you can connect only files that are committed and pushed to git when running remotely, which is exactly what i'm trying to find a workaround to.

  
  
Posted 4 months ago

@<1523701070390366208:profile|CostlyOstrich36> Hi, i would expect a feature that looks something like this:
clearml Task CLI option "--mount-files" (or other informative name) which would be used to add local files to the sent task in the following format:

clearml-task --project examples --name remote_test --script my_script.py --mount-files "local_file_1:target_path_1, local_file_2:target_path_2"

of course there would be some size limit to the mounted files (same as you do with limiting uncommitted changes size).

needles to say that the expected behavior is to get a copy of the specified files in the specified target paths.
I think it is an easy to add feature and users could benefit quite a lot from it. what do you think?

  
  
Posted 4 months ago

Hi,
If your config file is a YAML or JSON or something similar, you can just connect it to your task using task.connect_configuration(dict) after loading it in your script. Then your full configuration can be seen in ClearML and retrieved in your script using task.get_configuration_object_as_dict()
see None

  
  
Posted 4 months ago

Hi @<1594863230964994048:profile|DangerousBee35> , I don't think there is such a mechanism currently. What would the expected/optimal behaviour would be in your use case?

  
  
Posted 4 months ago
561 Views
4 Answers
4 months ago
4 months ago
Tags
Similar posts