Forum Discussion

numpy's avatar
numpy
Frequent Visitor
7 years ago
Solved

Dynamic Measure Calculation Power BI DAX

Hi All,

 

I am like just getting started with power bi and am real newbie to DAX.

 

While I was exploring Power BI I came across a question/problem. Whether the measures can have dynamic calculations,

 

Problem :

 

 

 

 

 

 

 

Above is made up data for explaining the problem.

 

I want a measure to dynaically calculate % with respect to selection made in slicer.

 

Slicer 1: Category selection

Slicer 2 : Orange , Red , blue , green(mange to get it into slicer by doing some secondary table)

 

Calculation needed:

 

If Orange selected in slicer 2 measure should be Sum(Orange)/Sum(Total Orange Action)

 

i.e if slicer 1 has clothes and food selected : (1+1)/ (10+12) = 0.0909

 

If Orange and Some other value(for example taking red) is selected measure should be sumthing like :

 

Sumx(Orange + Red)/ sumx(total orange action + total red action) 

 

i.e if slicer 1 has clothes and food selected : ((1+23)+(1+43))/ ((10+200)+(12+250))= 68/472 = 0.144

 

And so on if three value selected:

 

Sumx(all three value) / sumx(total three values action)

 

I am not sure whether this is possible or not and would appretiate any sort of help. I want to use this in a line chart where the user can select the measures and see the percentage against time.

Need Help
Tips and Tricks
  • Anonymous's avatar
    Anonymous
    7 years ago

    HI numpy,

     

    I'd like to suggest you do 'unpivot columns' on query editor site to convert your table to category, color, value, action color, action value.

    Unpivot columns (Power Query)

    Then you can simply use measure and slicer to achieve your requirement.

     

    Sample:

     

    Transformed query table

     

    Full query:

    let
    Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("NY69CoAwDITfJXOH/kVxFtzc3IqDoKBQKKj4/PbSdshxbb5cEgKNMb3n8ZAik8u6LB4GYjSc1s2yplUFmlLaK+/BS+FlLHgWnrN0hV/u7TsiWgAdktFl2zgvK1xZWWbm63lrNOO/R9wg51RKzqkD6w8=", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type text) meta [Serialized.Text = true]) in type table [Category = _t, orange = _t, red = _t, blue = _t, green = _t, #"Total Orange action" = _t, #"Total red action" = _t, #"Total blue action" = _t, #"Total green action" = _t]),
    #"Changed Type" = Table.TransformColumnTypes(Source,{{"Category", type text}, {"orange", Int64.Type}, {"red", Int64.Type}, {"blue", Int64.Type}, {"green", Int64.Type}, {"Total Orange action", Int64.Type}, {"Total red action", Int64.Type}, {"Total blue action", Int64.Type}, {"Total green action", Int64.Type}}),
    #"Unpivoted Columns" = Table.UnpivotOtherColumns(#"Changed Type", {"Category", "Total Orange action", "Total red action", "Total blue action", "Total green action"}, "Color", "Value"),
    #"Unpivoted Columns1" = Table.UnpivotOtherColumns(#"Unpivoted Columns", {"Category", "Color", "Value"}, "Attribute2", "Action"),
    #"Replaced Value"  = Table.ReplaceValue(Table.ReplaceValue(#"Unpivoted Columns1","Total ","",Replacer.ReplaceText,{"Attribute2"})," action","",Replacer.ReplaceText,{"Attribute2"}),
    #"Filtered Rows" = Table.SelectRows(#"Replaced Value", each [Color] = Text.Lower([Attribute2])),
    #"Removed Columns" = Table.RemoveColumns(#"Filtered Rows",{"Attribute2"})
in
    #"Removed Columns"

    Measure formula:

    Percent = 
CALCULATE (
    SUM ( 'Sample'[Value] ),
    VALUES ( 'Sample'[Category] ),
    VALUES ( 'Sample'[Color] )
)
    / CALCULATE (
        SUM ( 'Sample'[Action] ),
        VALUES ( 'Sample'[Category] ),
        VALUES ( 'Sample'[Color] )
    )

     

    Regards,

    Xiaoxin Sheng

3 Replies

  • Anonymous's avatar
    Anonymous
    Not applicable
    7 years ago

    HI numpy,

     

    I'd like to suggest you do 'unpivot columns' on query editor site to convert your table to category, color, value, action color, action value.

    Unpivot columns (Power Query)

    Then you can simply use measure and slicer to achieve your requirement.

     

    Sample:

     

    Transformed query table

     

    Full query:

    let
    Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("NY69CoAwDITfJXOH/kVxFtzc3IqDoKBQKKj4/PbSdshxbb5cEgKNMb3n8ZAik8u6LB4GYjSc1s2yplUFmlLaK+/BS+FlLHgWnrN0hV/u7TsiWgAdktFl2zgvK1xZWWbm63lrNOO/R9wg51RKzqkD6w8=", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type text) meta [Serialized.Text = true]) in type table [Category = _t, orange = _t, red = _t, blue = _t, green = _t, #"Total Orange action" = _t, #"Total red action" = _t, #"Total blue action" = _t, #"Total green action" = _t]),
    #"Changed Type" = Table.TransformColumnTypes(Source,{{"Category", type text}, {"orange", Int64.Type}, {"red", Int64.Type}, {"blue", Int64.Type}, {"green", Int64.Type}, {"Total Orange action", Int64.Type}, {"Total red action", Int64.Type}, {"Total blue action", Int64.Type}, {"Total green action", Int64.Type}}),
    #"Unpivoted Columns" = Table.UnpivotOtherColumns(#"Changed Type", {"Category", "Total Orange action", "Total red action", "Total blue action", "Total green action"}, "Color", "Value"),
    #"Unpivoted Columns1" = Table.UnpivotOtherColumns(#"Unpivoted Columns", {"Category", "Color", "Value"}, "Attribute2", "Action"),
    #"Replaced Value"  = Table.ReplaceValue(Table.ReplaceValue(#"Unpivoted Columns1","Total ","",Replacer.ReplaceText,{"Attribute2"})," action","",Replacer.ReplaceText,{"Attribute2"}),
    #"Filtered Rows" = Table.SelectRows(#"Replaced Value", each [Color] = Text.Lower([Attribute2])),
    #"Removed Columns" = Table.RemoveColumns(#"Filtered Rows",{"Attribute2"})
in
    #"Removed Columns"

    Measure formula:

    Percent = 
CALCULATE (
    SUM ( 'Sample'[Value] ),
    VALUES ( 'Sample'[Category] ),
    VALUES ( 'Sample'[Color] )
)
    / CALCULATE (
        SUM ( 'Sample'[Action] ),
        VALUES ( 'Sample'[Category] ),
        VALUES ( 'Sample'[Color] )
    )

     

    Regards,

    Xiaoxin Sheng

    • numpy's avatar
      numpy
      Frequent Visitor
      7 years ago
      Thanks alot. To be honest i didn't think of this. Very helpful.