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
Hey All, I Am Reporting A

Hey all,
I am reporting a DataFrame using report_table after finishing some model testing and like to compare between two DF from two different tasks,
When going to compare view, I am seeing those tables in the Plots tab side by side, is there a way to compare and see the diff between them?
Maybe report something else and not table?

Thanks:)

  
  
Posted 6 months ago
Votes Newest

Answers 4


@<1523701087100473344:profile|SuccessfulKoala55> Thanks, you mean downloading the previous from clearML server and make the diff and report it?
It can be a solution but less dynamically since lets say I Iike to change my baseline it will cause to run all the tests again…

  
  
Posted 6 months ago

Hey, @<1523701070390366208:profile|CostlyOstrich36>
They both should have the same values and I like to see where they not match.
So even basic stuff like df1==df2 , for example:

>> import pandas as pd
>> df = pd.DataFrame({'a': ['axy a', 'xyz b'], 'b': ['obj e', 'oaw r']})
>> df2 = pd.DataFrame({'a': ['ax2 a', 'xyz b'], 'b': ['obj e', 'oaw r']})

print(df == df2)
       a     b
0  False  True
1   True  True
  
  
Posted 6 months ago

Hi @<1523701977094033408:profile|FriendlyElk26> , let's say you have a table, which you report. How would you suggest comparing between two tables?

  
  
Posted 6 months ago

@<1523701977094033408:profile|FriendlyElk26> you can always report the diff from the previous table?

  
  
Posted 6 months ago