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
I Try To Update Scalars, But The Iteration I Pass Is Ignored, Instead The Last Iteration Is Incremented, Feels Wrong To Me.

I try to update scalars, but the iteration I pass is ignored, instead the last iteration is incremented, feels wrong to me.
prev_task = Task.init(continue_last_task=id)
prev_task.get_logger().report_scalar("model", "accuracy", 2, 0)

  
  
Posted 3 years ago
Votes Newest

Answers 5


Hi NastyOtter17 , I'm not sure I understand - can you explain what you see in the UI after running this as opposed to what you expect to see?

  
  
Posted 3 years ago

This seems to be in line with what you see

  
  
Posted 3 years ago

Indeed, does what stated in the docu, however I think its a bit odd, as .report_scalar() works quite different in this case compared to the normal case and iteration is not an optional param but will be ignored anyway

  
  
Posted 3 years ago

Sure!
Before:
{'model': {'accuracy': {'name': 'accuracy', 'x': [0, 1], 'y': [0.5789473652839661, 1.0]}}
After:
{'model': {'accuracy': {'name': 'accuracy', 'x': [0, 1, 2 ], 'y': [0.5789473652839661, 1.0, 2.0 ]}}
Expected:
{'model': {'accuracy': {'name': 'accuracy', 'x': [ 0, 1], 'y': [ 2.0 , 1.0]}}

  
  
Posted 3 years ago

NastyOtter17 from Task.init() 's docstring regarding continue_last_task :
` continue_last_task (bool ) – Continue the execution of a previously executed Task (experiment)

When continuing the executing of a previously executed Task, all previous artifacts / models/ logs are intact. New logs will continue iteration/step based on the previous-execution maximum iteration value. For example: The last train/loss scalar reported was iteration 100, the next report will be iteration 101. `

  
  
Posted 3 years ago
608 Views
5 Answers
3 years ago
one year ago
Tags