we can run the job without using the removedublicates stage. In the sort stage we have one option that is allow dublicates=false,if you set allow dublicates=true,dublicate records also moves from source to target.
By using aggregator stage and transformer stage. Use the aggregator stage and group the column. In the transformer stage, use if then else condition and count(column)>1 then duplicates else no duplicates.
By using the merge stage we can remove the dublicates of the source data.Target will capture the unique records of the source, It won't allow the dubilcates