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
Hi, Does Anyone Know That When Using Tensorboard To Record Loss (Such As

Hi, Does anyone know that when using tensorboard to record loss (such as writer.add_scalar('Loss/train', loss.data, epoch*len(trainLoader)+step) ), where will the data be stored? elastic or mongo db? Can I change its storage location? Thanks in advance 🙏

Posted 3 years ago
Votes Newest

Answers 5

Hi FierceFly22 🙂 ,
When reporting a scalar, the value will be stored as an entry in a specific Elastic index by the server. Why do you need to change the storage location? What is the use-case?

Posted 3 years ago

Hi SuccessfulKoala55 Because of security concerns, I would like to upload these scalars to AWS, just like we upload artifacts use output_uri . Do you have any suggestions?

Posted 3 years ago

Well, if you run the Trains Server in AWS EC2, the Elastic instance the server uses will run in your machine in AWS - will that suit your requirement?

Posted 3 years ago

Got it. Let me study how to do it, thank you for your reply 😄

Posted 3 years ago

Sure 🙂

Posted 3 years ago
5 Answers
3 years ago
one year ago