lets say your table called table_1 has the following fields
(col_1, col_2,col_3)
step 1: as a safety measure back table_1 to table_backup by issuing
create table_backup as select * from table_1;
step 2: remove all non duplicate rows to another new table
create table_2 as select distinct col_1, col_2,col_3 from table_1;
step 3: delete all rows from table_1 by issuing
delete * from table_1;
commit;
step 4: bring back all rows from table_2 to table_1
insert into table_1 select * from table_2;
step 4: you should not have duplicated in table_1 now, you can delete table_2 and table_backup
drop table_2;
drop table_backup;
---------------------------------
another quick not safe way would be
delete from table_1 where col_1 not in
select distinct col_1from table_1;
commit;
//here col_1 will be the column you use to see there are duplicares