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

Hi 👋
How can I connect clearml to a relational database, and have sql query as a dataset? (e.g. dataset.add_references(query = “select * from images where label = ‘1’)). I would love clearml to reference to the data obtained through that query, without loading the data into clearml itself.
Can this be done somehow?

Thank you!
Cheers,
Valentin

  
  
Posted 6 months ago
Votes Newest

Answers


Hi @<1695969549783928832:profile|ObedientTurkey46>

How can I connect clearml to a relational database, and have sql query as a dataset? (e.g. dataset.add_references(query = “select * from images where label = ‘1’)).

hmm interesting, you have a couple of options that I can think of:

  • You can have your query and an argument to the Task, which means it is logged and can be changed later from the UI when you are relaunching it.
  • You can have the query an an argument for a preprocessing Task that creates a dataset, then pass the dataset ID (i.e. and all data) as argument for any Tasks that needs to use that data
  • And you can create a pipeline where the query is part of the pipeline and the data is passed between the components (wither as a dataset or as pandas object). wdyt?
  
  
Posted 6 months ago