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
Can Anyone Help Me To Explain The Differences Between

Can anyone help me to explain the differences between path_substitution and direct_access . As I understand:

  • direct_access : register data will not be pushed to ClearML server.Dataset.create then upload => Server only save the local path of the file.
    Dataset.get => server will return the local path url
  • path_substitution : is it like symlink from local link to fake clearml_server link. I still don't get the idea how the server and local user interact hereIs my understanding correct? Can anyone help me to clarify this?
Posted 13 days ago
Votes Newest

Answers 3

Hi @<1570583237065969664:profile|AdorableCrocodile14> , I think this is a good answer: None

Posted 13 days ago

@<1523701087100473344:profile|SuccessfulKoala55> Is there any way .get_local_copy return direct link of those local files items. Data is duplicated. Those data existed in local storage, but they are still needed to save in cached folder

Posted 3 days ago

can you give me an example of use of direct_access setup 🙇

I have large dataset, would like to register the dataset in ClearML without uploading real dataset. When running a task, I can get the dataset without creating a cache folder

Posted 3 days ago