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
Hello, I Have Questions Relating To Metric Storage On The Web App. We Recently Crossed 1.2 Gigs Of Free Metric Storage And We Were Looking To Reduce That Figure To Stay In The Free Tier. We'Ve Deleted Thousands Of Experiments But It Doesn'T Seem To Have

Hello,

I have questions relating to metric storage on the web app. We recently crossed 1.2 gigs of free metric storage and we were looking to reduce that figure to stay in the free tier. We've deleted thousands of experiments but it doesn't seem to have changed the amount of metric stored reported. I also find it's a little strange that we've been under 1.2 gigs for over a year and suddenly we're at over 12 gigs. I find it a bid hard to track down what's directly contributed to this excessive storage. I suspect it's related to some recent hyperparameter sweeps but we deleted the related experiments in the project but it's still reporting quite high. How do you suggest we go about assessing what experiments correspond to the higher metric storage?

  
  
Posted one month ago
pic
More
EnthusiasticCow4
0 × 1
Report
Votes Newest

Answers 3

Hi @<1523701087100473344:profile|SuccessfulKoala55> - We tried to delete some additional hyperparameter tunings but it doesn't seem to have impacted metrics stored. It's not clear to me what is occupying all the metric storage space.

  
  
Posted one month ago
pic
More
EnthusiasticCow4
0 × 1
Report

Hi @<1545216070686609408:profile|EnthusiasticCow4> ,how long after deleting the data did you check the usage again?

  
  
Posted one month ago
pic
More
SuccessfulKoala55
0 × 1
Report

It did update but I'm beginning to see the issue now. It seems like the metrics on thousands of experiments amounted to a few MB where deleting one of the hyperparameter experiments freed up over a gig. I'm having difficulty seeing why one of these experiments occupies so much metric space. From the hyperparameter optimization dashboard the graphs and the tables might have a few thousand points.

Can you help me:

  • Better understanding why it's occupying so much metric storage space
  • Is there a better way to minimize that in the future?
    Thanks!
  
  
Posted one month ago
pic
More
EnthusiasticCow4
0 × 1
Report
Write your answer
173 Views
3 Answers
one month ago
one month ago
Tags
clearml