Auto ML Error

[Etsui]

Hi everyone,

Thanks in advance for your help. I am testing out Power BI's AutoML feature and have run into a problem when I am trying to train the model.

The error message I got was : "Expression.Error: We cannot convert the value "[Function]" to type Function".

Digger deeper in Advanced Editor, I believe the issue came from the 2nd last row.

let

  Source = #"Historical Data",

  #"Selected columns" = Table.SelectColumns(Source, {"BU Name", "COUNTRY", "GENDER", "CITIZENSHIP", "Intake Semester", "STUDY_PROG", "SCAGENT", "Converted Status"}),

  #"Removed nulls" = Table.SelectRows(#"Selected columns", each [Converted Status] <> null and [Converted Status] <> ""),

  #"Add conditional column" = Table.AddColumn(#"Removed nulls", "Converted", each if [Converted Status] = "Converted" then true else false),

  #"Changed Type" = Table.TransformColumnTypes(#"Add conditional column", {{"Converted", type logical}}),

  #"Remove column" = Table.RemoveColumns(#"Changed Type", {"Converted Status"}),

  #"Sampled input Wrapper" = (table) => AI.SampleStratifiedWithHoldout("Converted", Table.RowCount(table), table),

  #"Sampled input" = #"Sampled input Wrapper"(#"Remove column")

in

  #"Sampled input"

Being a complete novice, can someone let me know what it is intending to do in that row? I believe it's prompting me to provide some sort of table, but I am not sure.

Thank you.

E

[Anonymous]

Hi @Etsui 

It seems that you're hitting the PQ firewall. In order for this to work, make sure that the privacy level for both data sources is compatible (Public/Public or Organizational/Organizational), or disable the firewall through the "Fast Combine" option.

The issue is that the firewall can't buffer certain things like functions.

You may refer to this blog to set Fast Combine options.

Best Regards,

Rico Zhou

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.