how to find NaN values of each row of spark dataframe to decide whether the rows is dropeed or not

[muhammet pakyürek]

is there any way to do this directly.  if its not, is there any todo this 
indirectly using another datastrcutures of spark