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
I Was Working With Pipelines And Just Wanted To Know If Using Clearml Data Is A Better Way To Store Any Task Outputs Compared To Artifacts. Also For Example If I Have 3 Tasks -> Preprocessing -> Training -> Eval. Is It Better To Store The Output Model In

I was working with pipelines and just wanted to know if using clearml data is a better way to store any task outputs compared to artifacts. Also for example if i have 3 tasks -> Preprocessing -> training -> eval. Is it better to store the output model in clearml data.

  
  
Posted 7 days ago
Votes Newest

Answers


Hi @<1859043976472956928:profile|UpsetWhale84> , output models are registered in the model repository.

It really depends on the use case regarding artifacts vs datasets. If the outputs are relevant only in the context of the pipeline then use artifacts, otherwise use datasets

  
  
Posted 6 days ago
184 Views
1 Answer
7 days ago
6 days ago
Tags