It would be useful to have an M function to keep duplicates based on selected column(s). This would allow the users to see the user to recognise where the duplicates occur and to choose which row to keep in the Table (from the duplicates). At this moment to find the duplicates I am using GroupBy, Allrows, Count rows, Select the groups that has more than 2 in count. For large dataset that has over 500000 rows having few duplicates, this approach creates unwanted too many groups a Thus, this is not an efficient method I feel. A Table.NonDistinct function in M may help. Is that possible?
2 Comments
- Maybe this option would be useful: View -> Data Preview -> Column profile -> You can recognize values with duplicates.
- Status changed:NewtoAbandoned
We actually have an option out of the box today. You can find it inside of the Home tab inside the "Reduce rows" group, and with the label "Keep duplicates" as shown in the screenshot below
