The final cumulative sum of every row, is in fact simply the sum of every row, or the row-wise sum, so we can implement this as:
>>> x.sum(axis=1)
array([ 10, 35, 60, 85, 110])
So here for every row, we calculate the sum of all the columns. We thus do not need to first generate the sums in between (well these will likely be stored in an accumulator in numpy), but not "emitted" in the array.
