groups date/hours

Hi Anonymous 

If I understand correctly, you have too many entries, so need to "compress" rows, here is one way with the assumption you do have the entry for each 10 minute, paste in Advanced Editor

let
    Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("ldI/D4IwEIfhr2I6ixztlX8rg3E1ukiIYdGYOPr9Y5FLwXDl6Eqe0ML7a1v1aF63vnmfPle1V2BSyFMNOttBVmfgHhVadftVpkd2yJZwIAnkCZgfrGogy7w0aE2ExQhrI2weYQvphxmCpQSRYCWdjgWdLv6rSXK9/o+3BMVYtvIvFS6gHfRWiOUseivEcnb6sDFW/4T8cj+ei9CwDZVagZpguYSB/IZibbG4XVKsbVZHWBNhmftyw0aKtQKRIFMqMFfkYgUt04ubNsqx/LSZbwoM20qxZsO2UqzZsK0Uy1l/2yFW9wU=", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [item = _t, dt = _t, value = _t]),
    #"Changed Type" = Table.TransformColumnTypes(Source,{{"item", type text}, {"dt", type datetime}, {"value", type number}}),
    #"Sorted Rows" = Table.Sort(#"Changed Type",{ {"item", Order.Ascending},{"dt", Order.Ascending}}),
    #"Added Custom" = Table.AddColumn(#"Sorted Rows", "group", each Number.From( DateTime.Date([dt]))*100+ Time.Hour([dt])*10+  Number.RoundDown( Time.Minute([dt])/10)),
    #"Grouped Rows" = Table.Group(#"Added Custom", {"item", "group"}, {{"minDT", each List.Min([dt]), type nullable datetime}, {"avgValue", each List.Average([value]), type nullable number}}),
    #"Removed Columns" = Table.RemoveColumns(#"Grouped Rows",{"group"})
in
    #"Removed Columns"

Hi Anonymous ,

Just add the corresponding columns when grouping tables:

Best Regards,
Community Support Team _ Yingjie Li
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.