I need some help w/a programming project...

Okay...don't laugh...

I am a freshman C.S. major at my college and I am starting out learning c++ on a linux console machine. We are currently writing a program that reads in a list of numbers from a file and populates an array with those values. Then, it sorts the values and computes the mean, median, mode, and standard deviation. I am having one slight problem. The function I wrote for finding the mode only returns the highest value(ie, if there are two or three modes it only returns one of them - the highest value). I have included my humble little piece of code and if it isn't too much trouble with for you gurus I would greatly appreciate some assistance...here are the questions I need answered:

1. Is there a way to simply modify my existing code(i.e. changing the type to void and adding another loop over the top)

2. Do I need to use another array to determine the frequency of each value and if so, how do I implement it?

3. If I do need to use an array for frequency, how could I compare the two arrays to find all of the correct frequency indices?

If I am posting in the wrong forum, sorry...if you know where I could get help on this I would love to know...anyway...here is my function for the mode:

(I realize I will probably have to change its type to void since I need to find multiple modes...as it is, it returns the highest value mode regardless of how many there are)

Code:

---

int mode(int array[], int SIZE_OF_ARRAY)
{

        using namespace std;

        int counter = 0;
        int max = 0;
        int i;
        int mode_index = 0;

        for (int i = 0; i < SIZE_OF_ARRAY; i++)
        {
                if (array[i] == array[i + 1])
                {
                        counter++;
                }

                else
                {
                        if (counter > max)
                        {
                                max = counter;
                        }

                        counter = 0;
                }
        }       

        counter = 0;


        for (i = 0; i < SIZE_OF_ARRAY; i++)
        {
                if (array[i] == array[i + 1])
                {
                        counter++;
                }

                else
                {
                        if (counter == max)
                        {
                                mode_index = i;
                        }

                        counter = 0;
                }

        }

        return array[mode_index];

}

---

You have so many options here. But we need some info first. First, do you know any of the STL? Did you learn vector<> or list<> or any of that good stuff? Second, why are you returning a double when your array is an array of int? Third, when pasting code, try using the code tags given in the forum. There's a button that says "PHP" at the vB Code section of the reply. It helps if you put your code in there.

Welcome to the code guru forums! Though I know this specific topic isn't already in this forum (it's too specific), please be aware of the fact that there's a FAQ and that there is a search forum feature -- you can often find what you need there :)

lol...don't know why my function is type double...i'll change that...doesn't hurt anything I guess but thanks for pointing that out...

I haven't learned much as this is an intro to C.S. class and mostly focuses on C++ programming basics. vector<> and list<> are unfamiliar to me. Also, I will edit my post above using the code tags. Anyway...I think my teacher wouldn't be too impressed if I used stuff that we haven't learned. I was hoping for a solution involving one of these options:

1. Maybe a void function that fills an array of the same size as the sorted array of values that has frequency numbers like...

value array frequency array
21___________3
21___________3
21___________3
35___________1
44___________2
44___________2
50___________3
50___________3
50___________3

...and then another void function that outputs the value array indices that have the max frequency value(in the above example it would output array[0] and array [6] or 21 and 50)

2. Adding another loop to the existing code that does iterations excluding the already found max until all of the same maxes are found(i.e. if there are two modes of frequency 4, it couts the first mode index then iterates without including those particular indices and eliminates them one at a time until the loop terminates)

i'm definitively not a big fan of C++, but i can tell you that you're not programming in C++, that looks like C to me.

My textbook says I am programming in C++...we are not learning anything fancy...like i said...this is an intro to C.S. class...most people in my class don't even type 20 wpm...just in case you were curious my textbook is Problem Solving with C++ The Object of Programming, 4th Edition by Walter Savitch.

Quote:

---

Originally posted by Luis G
that looks like C to me.

---

notice that my counter is incremented with ++?

You are using C, not C++... but...

I have modified your function:

void mode(int array[], int SIZE_OF_ARRAY, int* numMode)
{
//using namespace std;
int count = 0;
int max = 0;
int* data = (int*)malloc(SIZE_OF_ARRAY*sizeof(int));
int count2 = 0;

for(int i = 0; i < SIZE_OF_ARRAY; i++)
{
count = 0;
for(int j = i; j < SIZE_OF_ARRAY; j++)
{
if(array[i] == array[j])
{
count++;
}
}
if(count > max)
{
max = count;
count2 = 0;
*(data + count2) = array[i];
}
else if(count == max)
{
count2++;
*(data + count2) = array[i];
}
}

count2++;
for(i = 0; i < count2; i++)
{
array[i] = *(data + i);
}

*numMode = count2;
free(data);
}

When finishing, numMode returns the number of elements of the mode, and are placed in the first positions of the array.

i.e. int array[] = 9, 3, 7, 10, 12, 9, 3, 10;
int numElem;
mode(array, 8, &numElem);

numElem will be 3, and array[0-2] will be 9, 3, and 10

I would do something along these lines then if you have to stay with basic types. I'm doing this on the fly so I might make a stupid syntax error or have a brain fart, but I hope you get the idea :)

Code:

---

int mode(int array[], int SIZE_OF_ARRAY, int *pModeArray)
{
int ModeArrayCounter = 0;

... blah blah your code here

                        if (counter > max)
                        {
                                max = counter;
                                ModeArrayCounter = 1;
                        }
                        else if(counter == max)
                        {
                                ModeArrayCounter++;
                        }
... more code here
}

if(ModeArrayCounter > 0)
{
int Current = 0;
// create a new array using your pointer
// ModeArrayCounter in size, so it can hold
// all of the items that match that mode
pModeArray = new int[ModeArrayCounter];
... your for loop code here
if(counter == max)
{
*pModeArray[Current] = array[i];
Current++;
}
... more code
}

return ModeArrayCounter; // return the size of the new array pModeArray so you can loop through it later.
}

---

ok I hope I made some sense here, I typed in intervals cuz my boss kept walking by :D

Quote:

---

Originally posted by xsyntricity
My textbook says I am programming in C++...we are not learning anything fancy...like i said...this is an intro to C.S. class...most people in my class don't even type 20 wpm...just in case you were curious my textbook is Problem Solving with C++ The Object of Programming, 4th Edition by Walter Savitch.



notice that my counter is incremented with ++?

---

I should have been more clear, 'cause "that looks like C to me" is a little vague, I should have said, that is C code.

You'd be surprised by the ammount of books and teachers that confuse C with C++.

btw, increments a la ++ are part of the C language, and they are not distinctive of each language. C++ was named like that because it is "C increased", or at least that's what i read.

Quote:

---

Originally posted by Caronte

int* numMode)
int* data = (int*)malloc(SIZE_OF_ARRAY*sizeof(int));
*(data + count2) = array[i];
*(data + count2) = array[i];
*numMode = count2;
free(data);

---

I am not familiar with these operations...I am sorry if I am being a hassel but I should proably only use operations my class has learned thus far...is there a very beginner level, baisc way to do this? The way that was coded looked something similar to using call by ref params...is there a way to do it using by ref params instead of that '*' thing? Even if it involves a huge amount of code it would probably be better than doing something my teacher knows I am not familiar with...

See if my code makes any more sense. My approach, using the new operator and such, is the c++ method, where malloc and free is the straight c method. If you say it's a c++ class then I'm sure new/delete is the way you want to go.

Quote:

---

Originally posted by xsyntricity
My textbook says I am programming in C++...we are not learning anything fancy...like i said...this is an intro to C.S. class...most people in my class don't even type 20 wpm

---

Believe it or not, you would have better off learning this the real C++ way, usage of vectors<> not arrays. Then you wouldn't have to lug that size argument to your mode function. I don't understand why these concepts are not introduced first. Beginning students pick up the vector<> syntax and have less trouble than when a beginning student gets arrays and pointers thrown at them. I just don't get it sometimes with the way this language is being taught to beginners

Quote:

---

...just in case you were curious my textbook is Problem Solving with C++ The Object of Programming, 4th Edition by Walter Savitch.

---

Here is a review of your textbook:

http://www.accu.org/bookreviews/publ.../p/p001032.htm

You are already handicapped by a book that is not recommended by professional C++ programmers.

Quote:

---

notice that my counter is incremented with ++?

---

What does that have to do with C or C++? Incrementing an integer using ++ exists in 'C' also.

Regards,

Paul McKenzie

Quote:

---

Originally posted by Luis G
You'd be surprised by the ammount of books and teachers that confuse C with C++.

---

The review of the book that I posted a link to states that the book is "C with classes". Oh-oh. I guess there is no mention of usage of the standard C++ library, or anything that C++ has had to offer for the past, say, 7 or so years.

The book used by xsyntricity is obsolete (published 1996) and doesn't reflect the way that proper, modern, C++ should be taught and programmed. I will boldly claim that if this book were up to snuff, the assignment would have been easier to do, and not as difficult as it has become. Too bad.

Regards,

Paul McKenzie

Why does this keep happening? Why is c++ taught in c style to this day?? Why the misguidance for all these future job seekers?

xsyntricity, please, please tell your professor, in the nicest way possible so as not to trigger a territorial anger, that teaching c++ like this is not recommended by the c++ community, is not recommended by Bjarne Stroustrup himself, and really, truly causes problems for the students when they hit the workplace. I hope some of the posts in this thread have helped you solve your original problem, and I strongly urge you if you really have an interest in c++ to check out the ACCU site posted by Paul McKenzie and get a couple of the books that are recommended there.

None of this is to say that c is not a beautiful language in itself, and is used heavily in the marketplace to this day. But good c style is not good c++ style...

Quote:

---

Originally posted by Paul McKenzie
Believe it or not, you would have better off learning this the real C++ way, usage of vectors<> not arrays. Then you wouldn't have to lug that size argument to your mode function. I don't understand why these concepts are not introduced first. Beginning students pick up the vector<> syntax and have less trouble than when a beginning student gets arrays and pointers thrown at them. I just don't get it sometimes with the way this language is being taught to beginners.

You are already handicapped by a book that is not recommended by professional C++ programmers.
What does that have to do with C or C++? Incrementing an integer using ++ exists in 'C' also.

Regards,

Paul McKenzie

---

Well, first of all, I did not select our textbook and have thus been learning all of this the hard way I guess...I am not surprised to see a bad review of the text as it is more geared for a one-time experience with the 'C' programming language and doesn't seem suited for a C.S. major class - even an intro; however, I have no choice but to do this the way my instructor and textbook have taught it...so...I am stuck between a rock and a hard place...again...I am sorry for being such a hassel and now that you know what a hinderance my textbook has been I hope you understand my situation a little better...

Here is a quote from my text:

"Most of C is a subset of C++ , and so most C programs are also C++ programs."

This is why I was under the assumption I was programming in C++...according to my text, most C code is C++ code...although the opposite is not true...seeing as how the title to my text begins with C++, a beginner such as myself would naturally assume they were learning C++...sorry for the mistake...now that that is out of the way and we are both on the same page...is there any way to get this done without using vectors or '*'s? I have basic knowledge of functions, types, loops, nested loops, overloading function names, call by ref params, streams and basic file i/o, boolean expressions, arrays, and multi-dimensional arrays...I understand that there is a simple, easy way to do this...however, we have taken to learning the hard way first so I must do this the hard way...any other help you can give me would be much appreciated...

whoops

Hey. nonsense. I myself was a college teacher (not in CS thank god) and have to tell you that any Prof. worth a s**t would be delighted to see a student learning stuff on there own and utilizing it to solve problems in a more elegant way. The key is that you have to be able to explain what you did and why you did it. Galathaea has given a good reason why and as for what: Make sure you really understand the solution given. If you are going to tell me that you have a Professor who will frown on learning then I am going to tell you BAH :mad:

Now, the question is are you a good student. Do you want to take the time to learn a little STL so you can solve this problem the C++ way?

Quote:

---

Originally posted by xsyntricity
Well, first of all, I did not select our textbook and have thus been learning all of this the hard way I guess...I am not surprised to see a bad review of the text as it is more geared for a one-time experience with the 'C' programming language and doesn't seem suited for a C.S. major class - even an intro; however, I have no choice but to do this the way my instructor and textbook have taught it...so...I am stuck between a rock and a hard place...again...I am sorry for being such a hassel and now that you know what a hinderance my textbook has been I hope you understand my situation a little better...

---

That's OK, since you wouldn't have known before-hand what you should have expected.

Quote:

---

Here is a quote from my text:

"Most of C is a subset of C++ , and so most C programs are also C++ programs."

---

This is not entirely true. There are gobs of 'C' programs that cannot compile under C++. Here is one such example:

Code:

---

int main()
{
    int class = 1;
    int template  = 2;
}

---

I hope you see why this would compile perfectly good 'C' code, but would make the C++ compiler throw up errors all over the place. What this should say is that "you can get a 'C' program to compile under a C++ compiler" and not that "most C is a subset of C++"

Quote:

---

This is why I was under the assumption I was programming in C++

---

The problem is that many concepts that a C programmer would learn are not the same concepts that a C++ programmer should learn. Basically C and C++ are two separate languages, where each has ways of doing things properly. When you get to the chapter on handling strings, how does the author of your book explain how to do this? (I shudder to see what is in there concerning string handling)

Quote:

---

is there any way to get this done without using vectors or '*'s?

---

The '*' is a pointer. When you deal with arrays, pointers are brought into play, whether you officially learned them or not. This is another reason why vector<> is easier -- you wouldn't have to know squat about pointers. But since you are using arrays, pointer knowledge is part of the deal. It is harder to "dumb-down" the code enough to not use pointers. So your best bet is to understand them now -- you're using arrays anyway.

Regards,

Paul McKenzie

Did my post not make sense to you either? I tried to keep it simple like your class. I was a TA for a while so I can relate to "having to use the basics" I thought I kept it pretty simple :confused:

let me know if you need me to clarify.

My ears and eyes are wide open...I have 'till tomorrow morning...i am using the g++ compiler(not sure what version) on a standard linux box...if you can explain how to make this work using STL i would be delighted to find out as I am a C.S. major and hope to be doing this stuff for a living some day. However, if I don't have a good explaination for why I did what I did my prof. won't be too impressed...so if you guys really feel like teaching a little beginner C++ I will stay here at my computer in this forum...I have all night...

;-)

Quote:

---

Originally posted by souldog
Hey. nonsense. I myself was a college teacher and have to tell you that any Prof. worth a s**t would be delighted to see a student learning stuff on there own and utilizing it to solve problems in a more elegant way.

---

Hear, hear. I also taught programming in college, and nothing would be more delightful than seeing a beginning student learn on their own.

Quote:

---

The key is that you have to be able to explain what you did and why you did it.

---

Exactly. If you can explain what you did, that is all that is necessary. Believe it or not, this is a trick used on some "do my homework for me" posters (not xsyntricity, you know, the ones that show no work whatsoever, but wants one of us to come up with the answer). What would happen is that the most advanced programming concepts are used to solve the problem. The trick is that the student hands in the assignment, and the professor sees that there is no way a beginning student could have come up with the code. Then the student has to sit (or stand there) and explain what his/her code is doing.

Quote:

---

Galathaea has given a good reason why and as for what: Make sure you really understand the solution given. If you are going to tell me that you have a Professor who will frown on learning then I am going to tell you BAH :mad:

---

Yes. As long as you can explain yourself, no professor should object to you learning on your own.

Regards,

Paul McKenzie

Eli...your post didn't make a whole lot of sense to me...sorry ;-)...as for pointers...i kind of thought that's what the * was but I didn't want to embarass myself more than I already have by guessing...also...if they are a part of arrays and would be the easiest way to do this i would love an explanation so that my prof. won't be angry i made use of them without us going over them...as for strings in my text...we haven't got that far but the subsections include an array type for strings, c-string variables, intro to the standard class "string", string processing with the class "string" and then it goes on to vectors...i see what you mean by this being a sh*tty text though...check this out...chapter 10 is arrays(what we are doing now), chapter 11 are strings and vectors, and chapter 12 are pointers and dynamic arrays...seems like it skips around a lot...also...if you could see the dependency chart you would probably throw up...

I guess the 1st order of business is what have you learned? You could create and array of structures where you'd have the value and a count ... That would give you all you'd need ... Then you can do your mathematical operations from there ...

also...i can show you my whole program if you want to...this isn't a "do my homework" thing as I already have a running program...just that one function doesn't do everything it needs to...i have included a typescript attachment of the code and the program running...

int* <-- you are correct, that is a pointer.

Check this out:

Code:

---

void CreateArray(int* pvar, int size)
{
pvar = new int[size];
}
void main()
{
int* IWantAnArray;
CreateArray(IWantAnArray, 10);
*IWantAnArray[0] = 1; // legal
}

---

ok so I passed in a pointer to an integer to a function. What the function did, then, was create a new array at the pointer address of pvar. Now, a pointer is pointing to an address. To get what's at that address, you use the * operator.

Code:

---

int* x;
int y = 4;
x = &y; // now pointing to the address of y
cout << *x << endl; // use * to get the value at the address (4)

---

So, since what's at my address in IWantAnArray is a newly created array, you use * to get at the value, and then you use standard array subscripting to get at the actual array, so [0], through [9] in this case.

Basically, I created an array on the fly of the exact size that you need for your code. You would pass in a variable of type int* just like I did here.

It's a shame that you can't use vector, cuz it would have been sooooo much easier :) I haven't used the method that I just explained in ages, so I hope it makes sense (and I hope I didn't F' it up :))

when you say "pointing" to the address, do you mean the memory address?

I would love to help! First, a little bit about vector. As you may have noticed, arrays are very difficult to work with and can be the source of errors since you always need to know how big it is, and that doesn't come packaged with the array. So you are always stuck passing around two values, which is not good c++ style because the whole goal of c++ is to encapsulate related information in things called objects and not have to reveal all the difficult pieces to clients (in simple terms, clients are functions that use the related information). Also, you probably have realized that arrays can never change size, and that is often frustrating and the source of errors.

Enter vector to the rescue! A vector is an array of data that can grow and keeps track of the size so you don't have to. To use it, you must first

Code:

---

#include <vector>

---

in the file that wants to use it.
Then you just make a vector instance like this

Code:

---

std::vector<int> myIntegerVector;

---

Where I have placed int, you put whatever type you want the array to be. It could be float, or some structure you have defined (structures and classes are types that you can define yourself -- you will find out about them in good time), or pointers, or all sorts of things. The strange <> brackets you will learn eventually have to do with templates, but again it is not something you have to worry about until you get there. You can use the vector without having to know anything about structures, classes, or templates. And the strange looking std:: before the name vector tells you the "namespace" where vector lives. I see you have

Code:

---

using namespace std;

---

in your code, so maybe you have learned something about namespaces. If not, you just need to know that the "using" keyword lets you use all of things living in the namespace that you have included without needing to type the std:: part. So you can actually just use

Code:

---

vector<int> myIntegerVector;

---

if you have the "using" line before the place you make your vector.

So my first suggestion is to get rid of your arrays (and the annoying size you have to keep passing around with it) and replace them with vectors. They are much safer and easier to use in the long run. And this would be a good time to see if you can get some information on the internet about how to use vector's, but don't worry, we can help you out with that as well. The first things you will want to be able to do is to set its size and be able to put some things into it. Here you will learn about using the "." in c++, which is the first step in using many of the wonderful things already lying around in the c++ language...

yeah, that's all a pointer does.

Consider this analogy:
int x;
consider int to be a house. When you do int x; you create a new house on a street somewhere.
int* y = &x; consider y to be the mailbox for the house. It's not the actual house, but it relates to the house -- it is the address of the house. If x didn't exist, then y would be useless in this example.

Anyway, I think your teacher's a little unrealistic. If you want to make your life easier, here's what I'd do:
pass in another array, the same size as your int array[].

Remove the "= new int[whatever]" code and just assume you have a second array the same size as your main array. Fill in the spots of the second array like you did the first array, and then return the integer just like you did before. Most of your second array will be unused, but oh well, if it's easier for you go that route.

I'd sit back and read all the posts (the ones that didn't contain the unnecessary arguements of C vs C++) and try them for a minute and see if anything clicks... try to do something on your own then come back to the forum if nothing is clicking.

good luck!

Quote:

---

Originally posted by xsyntricity
also...i can show you my whole program if you want to...this isn't a "do my homework" thing as I already have a running program...just that one function doesn't do everything it needs to...i have included a typescript attachment of the code and the program running...

---

Well, here is Caronte's code rewritten using vector:

Code:

---

#include <vector>
#include <iostream>

using namespace std;

void mode(const vector<int>& array, vector<int>& data)
{
    int count = 0;
    int max = 0;
    int nSize = array.size();
    data.resize( nSize );

    int count2 = 0;

    for(int i = 0; i < nSize; i++)
    {
        count = 0;
        for(int j = i; j < nSize; j++)
        {
            if(array[i] == array[j])
            {
                count++;
            }
        }
        if(count > max)
        {
            max = count;
            count2 = 0;
            data[count2] = array[i];
        }
        else if(count == max)
        {
            count2++;
            data[count2] = array[i];
        }
    }

    count2++;
    data.resize( count2 );
}

---

Note the absence of pointers in the mode function. There are also no hard-coded values that determine the maximum number of items to deal with. So you can have 1 item or 1,000,000 items, the mode() function will still work with no modification. The second parameter is another vector that will store the output -- this way the original vector that was passed in (the first argument) doesn't get overwritten.

So you can go with this if you want -- I didn't document it, because that is your job if you have to explain this to your professor.

Regards,

Paul McKenzie

If you go to this great page you will find Gabriel Fleseriu's beautiful introduction into using vectors in c++. It will be a great start for you!

well guys...i finally got it...and best of all...i didn't use any of you guys' code...just sat down and did it the hard way...so here it is...my mode function(with arrays instead of vectors and no pointers...i know, i know)

Code:

---

void mode(int array[], int SIZE_OF_ARRAY, int mode_array[]
        ,int& mode_array_size)
{

        using namespace std;

        int counter;
        int mode_value;
        int max = 0;
        int mode_index = 0;
        int frequency_array[SIZE_OF_ARRAY];

        for (int i = 0; i < SIZE_OF_ARRAY; i++)
        {
                counter = 0;

                for (int j = 0; j < SIZE_OF_ARRAY; j++)
                {
                        if (array[i] == array[j])
                        {
                                counter++;
                        }
                }

                frequency_array[i] = counter;

                if (counter > max)
                {
                        max = counter;
                }
        }

        mode_value = array[0];

        mode_array_size = 0;

        if (frequency_array[0] == max)
        {
                mode_array[mode_index++] = array[0];
                mode_array_size++;
        }

        for (int i = 1; i < SIZE_OF_ARRAY; i++)
        {
                if ((frequency_array[i] == max) && (array[i] != mode_value))
                {
                        mode_array[mode_index++] = array[i];
                        mode_array_size++;
                }


                mode_value = array[i];
        }


}

---

I have changed the code in order not to use '*' or mallocs or frees...
As in my previous code, the modes are returned in the first positions of array, while the number of modes, are returned in the "int" return of the function.


int mode(int array[], int SIZE_OF_ARRAY)
{
//using namespace std;
int count = 0;
int max = 0;
int count2 = 0;

for(int i = 0; i < SIZE_OF_ARRAY; i++)
{
count = 0;
for(int j = i; j < SIZE_OF_ARRAY; j++)
{
if(array[i] == array[j])
{
count++;
}
}
if(count > max)
{
max = count;
count2 = 0;
array[count2] = array[i];
}
else if(count == max)
{
count2++;
array[count2] = array[i];
}
}

count2++;
return count2;
}

thank you guys for all your help and advice...Caronte, thanks for the code and you time...

peace