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
Unanswered
Hi All. I Am Wondering How People Tend To Use Clearml With Cross-Validation. Do You Tend To Create Separate Experiments For Each Fold? And If So, Would You Then Create Another Experiment For The Aggregated Results?


Thanks AgitatedDove14 . I am not using ClearML for scheduling/execution at this stage. I am evaluating ClearML for adding reporting to our current workflow. We have existing (parallelised) code for cross-validating models and I am playing with how best to log training/testing to ClearML. One thought is to initialise a new clearML task in each fold to capture the iteration-level metrics, and then create another task/experiment at the end to capture the aggregated metrics across folds. Alternatively, I could simply dump all fold and aggregated metrics into a single experiment. I don't have a good feel yet as to the pros and cons and was wondering if anyone had any advice.

  
  
Posted one year ago
161 Views
0 Answers
one year ago
one year ago