我对我的输出感到困惑，该输出给出了用户输入的数字的总和的最接近平均值（将小数点四舍五入为整数）

Question

Complete the following program to find the closest number to average in the array. For example, the array {2, 4, 6, 3, 9, 10} has average of 5.666667 and closest number to average in the array is 6. Note that closest number can be larger than or smaller than the average.

#include <stdio.h>
#include <math.h>
int main()
{
    int i;
    double num[6],average, sum=0, closest;
    printf("Enter 6 doubles\n"); //2,4,6,9,3,10

    for (i=0; i<6; i++)
        scanf("%lf",&num[i]);

    //professor's code
    for (i=0; i<6; i++)
        sum += num[i]; //sum=34

    average=sum/6; //5.6666667
    closest = num[0]; //why are we even using this??? //initialize?

    for (i=0; i<6; i++)
        if(fabs(num[i]-average) < fabs(closest-average)) //|6-5.7|<|0-5.7|-> 0.7<5.7->true 
            closest = num[i]; // why is it giving 6?? 
   //where is it being rounded to the closest number??

    //professor's code ends     


    printf("Closest is %lf\n", closest);
    return(0);

}

//Output:
//Closest is 6

Answer 1

Your title and code comments suggest that you think the program (always) returns the whole number closest to the average of its inputs. It does not. It returns the input that is closest to the average of all the inputs. That will of course be a whole number if all the inputs are whole numbers, but you're inputting them as floating-point numbers, so they need not be whole numbers. The output therefore needn't be one, either.

For example, if these inputs are presented to your program:

1 2 2 2 2 1.8

, it will accept them and output

Closest is 1.800000

. Additionally, the output is not certain to be particularly close to the average. For example, try these inputs:

1 1 1 1 1 1000000

. You should get

Closest is 1.000000

. That's a whole number, but it misses the average by about 166666.

With respect to the questions embedded in the code:

  closest = num[0]; //why are we even using this??? //initialize? 

The loop immediately following that in the code looks for inputs that are closer to the average than closest is. That's meaningless if closest has not been assigned a value, and it potentially leads to the wrong result if closest does not contain a value equal to one of the inputs. Of course, with closest starting with the same value as num[0] , the output would be the same if the loop started testing at index 1 instead of index 0.

  //where is it being rounded to the closest number?? 

It is not rounded. At that point, closest already contains a value equal to one of the inputs. It does not need to be rounded, and in fact rounding would in many cases yield a result that was not among the inputs.