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
Question Regarding Tensorboard (If There Is An Answer Here Already Please Send Me A Link). I Have A Few Graphs With The Same X Axis But Different Y Axis That Are Presented On Different Graphs In Tensorboard And For Some Reason Trains Joins Them On The Sam

question regarding tensorboard (if there is an answer here already please send me a link).
i have a few graphs with the same x axis but different y axis that are presented on different graphs in tensorboard and for some reason Trains joins them on the same graph. is there a way to still view them on seperate graphs? (they are not in the same scale so it's not very useful for me to see them on the same graph)

  
  
Posted 4 years ago
Votes Newest

Answers 14


That will be sdk.metrics.tensorboard_single_series_per_graph

  
  
Posted 4 years ago

i updated the config file and it seems like its still combining the graphs. do i need to restart the server or something?

  
  
Posted 4 years ago

that would probably be the best 🙂

  
  
Posted 4 years ago

yes. im using 0.16.1

  
  
Posted 4 years ago

CloudyHamster42 you mean that when you set sdk.metrics.tensorboard_single_series_per_graph to True and you rerun the experiment, you are still getting multiple series on the same graph?
What's your Trains version?

  
  
Posted 4 years ago

CloudyHamster42 it will only affect news tasks created with the config file...sorry

  
  
Posted 4 years ago

I think there is a way, I'll have to check. BTW when you compare two tasks you do get separate graphs, right?

  
  
Posted 4 years ago

yes

  
  
Posted 4 years ago

Sounds odd, I bet there is a way to make it work witout implicit logging statement. This is with tf2 or pytorch? Which trains version are you using?

  
  
Posted 4 years ago

also its not working on new tasks, even when i add that to the config file

  
  
Posted 4 years ago

Cool. I found Logger.tensorboard_single_series_per_graph()

  
  
Posted 4 years ago

pytorch, trains version 0.16.0

  
  
Posted 4 years ago

Should work on new tasks of you use this command in the script. If you'd rather to keep the scripts as clean as possible, you can also configure it globally for all new tasks in trains.conf

  
  
Posted 4 years ago

BTW: CloudyHamster42 I think this issue was discussed on GitHub, and the final "verdict" was we should have an option to split/combine graphs on the UI side (i.e. similar to the "smoothing" or wall-time axis etc.)

  
  
Posted 4 years ago
945 Views
14 Answers
4 years ago
one year ago
Tags