Problem with leading zero's in a vector array of doubles

Question

Hi,

I'm tring to calculate the standard deviation of a vector of doubles (called A). Now I have a function called StDev that will do this. However, the first few elements of vector A are zero and I need to remove these. To do this I create a sub-array and then pass this to my StDev function as follows:

 std::vector<double> Array(find_if(Data.begin(), Data.end(), std::bind1st (std::not_equal_to<double>(), 0.0)), Data.end());     

 double standard_deviation = StDev(Array); 

Is there a way of doing this without having to create the vector 'Array' which is only used once. Can I somehow pass the required sub-array directly?

Thanks!

Answer 1

If you can modify your StDev function to take an iterator range instead of a whole container, you can do this quite easily:

template <typename ForwardIt>
std::iterator_traits<ForwardIt>::value_type 
StDev(ForwardIt first, ForwardIt last) { /* ... */ }

// called as:
double stdev = StDev(Data.begin(), Data.end());

// or:
double stdev = StDev(
    find_if(Data.begin(), Data.end(), 
            std::bind1st(std::not_equal_to<double>(), 0.0)), 
    Data.end()); 

Answer 2

You could change your StDev function so that it skips however many elements at the start of the array that are zero.