Some Funky Math

[BleaS]

Grader v. 1.01b.exe

Grader v. 1.01 BETA Source.zip

So, I have an odd dilemma that I'm not entirely certain I can pin point through debugging. To me, it seems more like an issue of not understanding how the system interprets a certain math equation than anything else.

To preface, my program is a basic grader program that I am building as a part of falling the Accelerated C++ book for learning C++. If you look at the code, you'll likely find inefficient uses of things, etc... however, the extent to what I know is what will be found in there, and suggesting that I replace "such and such" with a class will most likely confuse me at this point. Beyond that, if you know Accelerated C++ and know what I should know after reading Chapter 4, feel free to comment on the code and help as you will.

However, my issue lies in the calculation of final grades. At the moment, I have not incorporated a NULL state to my variables (I'm likely going to use the -1 method until I learn classes). Since I haven't, whenever you create a new student, it sets all of their basic attributes to 0, which you can then edit. However, whenever all of their attributes are 0, if I go to list to see their final grade, I do not see 0... I get odd numbers like "-1,#QHN", what the heck is that garbage?

The equation for what my program is doing to find the final grade is here:

double Do_Calc_Homework(Student_Record& student)
{
	sz_t Max_Homework = student.homework.size();
	double Total_Homework = 0;

	for (sz_t Curr_Homework = 0; Curr_Homework != Max_Homework; 
		++Curr_Homework)
	{
		Total_Homework += student.homework[Curr_Homework];
	}

	return Total_Homework / Max_Homework;
}

double Do_Calc_Quiz(Student_Record& student)
{
	sz_t Max_Quizzes = student.quizzes.size();
	double Total_Quizzes = 0;

	for (sz_t Curr_Quizzes = 0; Curr_Quizzes != Max_Quizzes; ++Curr_Quizzes)
	{
		Total_Quizzes += student.quizzes[Curr_Quizzes];
	}

	return Total_Quizzes / Max_Quizzes;
}

double Do_Calc_Test(Student_Record& student)
{
	sz_t Max_Tests = student.tests.size();
	double Total_Tests = 0;

	for (sz_t Curr_Tests = 0; Curr_Tests != Max_Tests; ++Curr_Tests)
	{
		Total_Tests += student.tests[Curr_Tests];
	}

	return Total_Tests / Max_Tests;
}

double Do_Calc_Project(Student_Record& student)
{
	sz_t Max_Projects = student.projects.size();
	double Total_Projects = 0;

	for (sz_t Curr_Projects = 0; Curr_Projects != Max_Projects;
		++Curr_Projects)
	{
		Total_Projects += student.projects[Curr_Projects];
	}

	return Total_Projects / Max_Projects;
}

double Do_Calc_Final(Student_Record& student, Weight& weight)
{
	return (weight.midterm * student.midterm)
		+ (weight.final * student.final)
		+ (weight.homework * student.homework_avg)
		+ (weight.tests * student.test_avg)
		+ (weight.quizzes * student.quiz_avg)
		+ (weight.projects * student.project_avg);
}

double Get_Average(const vector<double>& Number_List)
{
	double sum = 0;
	for (sz_t count = 0; count != Number_List.size(); ++count)
	{
		sum += Number_List[count];
	}

	double sum_size = Number_List.size();

	return sum / sum_size;
}

Any of the numbers divided by 0 should be 0. When the weights are multiple by 0, they should remains 0, the final grade should be 0, but it is not. Why? I'm pretty confused.

[laserlight]

Any of the numbers divided by 0 should be 0.

But you should not actually try to divide by zero. Rather, you should check if a division by zero will occur, and if so, return 0 (or throw an exception in other circumstances).

[Lindley]

Any of the numbers divided by 0 should be 0.

Any division by zero results in not-a-number (QNAN for double). So don't do that.

Second, you need an if statement in each of those loops ensuring you're finding the average of the positive values only----and that includes counting how many positive values there are rather than relying on .size()!

Of course, since you're using containers to store the various grades anyway, one wonders why you need a "no grade here" marker at all. Why not just expand the container only when you get valid grades?

[BleaS]

Any division by zero results in not-a-number (QNAN for double). So don't do that.

Second, you need an if statement in each of those loops ensuring you're finding the average of the positive values only----and that includes counting how many positive values there are rather than relying on .size()!

Of course, since you're using containers to store the various grades anyway, one wonders why you need a "no grade here" marker at all. Why not just expand the container only when you get valid grades?

It's impossible for a negative number to be entered into the value due to some other checks in the program, thus it's a bit easier since we don't have to check for negative numbers. Download the program and break it, I'm sure you can and it will teach me about what else I need to safeguard for.

I initialized values because of how it works when you add a student. Since the only way I found I could do push_back was to copy the entire container over, I needed to make a container in the function, and then push it back to the actual student_info container. I was under the assumption that in copying a structure container that I could not copy over uninitialized values, or, at the very least, that it would be bad form.

Thank you for the advice, I will relook at the calculating functions.

[STLDude]

Any of the numbers divided by 0 should be 0. When the weights are multiple by 0, they should remains 0, the final grade should be 0, but it is not. Why? I'm pretty confused.

I'm confused how could you even think like this. Dividing by 0 in math is invalid, so how could that be valid for computers?

[BleaS]

I'm confused how could you even think like this. Dividing by 0 in math is invalid, so how could that be valid for computers?

Because I was wrong.