Compute mean, median mode for map construct

[in_ship]

Hi guys!


I am writing code, which reads an input file (280 MB) containing a list of words. I would like to compute
1) total # of words
2) total # of unique/distinct words
3) mean/median/mode of word count (Link)

I managed to get done with 1) and 2), but my program crashes for 3). I am not quite sure whether there are memory issues or some bug in the code.

#include <iostream>
#include <fstream>
#include <string>
#include <map>
using namespace std;


double GetMedian(int daArray[], int iSize) {
    // Allocate an array of the same size and sort it.
    double* dpSorted = new double[iSize];
    for (int i = 0; i < iSize; ++i) {
        dpSorted[i] = daArray[i];
    }
    for (int i = iSize - 1; i > 0; --i) {
        for (int j = 0; j < i; ++j) {
            if (dpSorted[j] > dpSorted[j+1]) {
                double dTemp = dpSorted[j];
                dpSorted[j] = dpSorted[j+1];
                dpSorted[j+1] = dTemp;
            }
        }
    }

    // Middle or average of middle values in the sorted array.
    double dMedian = 0.0;
    if ((iSize % 2) == 0) {
        dMedian = (dpSorted[iSize/2] + dpSorted[(iSize/2) - 1])/2.0;
    } else {
        dMedian = dpSorted[iSize/2];
    }
    delete [] dpSorted;
    return dMedian;
}

double GetMode(int daArray[], int iSize) {
    // Allocate an int array of the same size to hold the
    // repetition count
    int* ipRepetition = new int[iSize];
    for (int i = 0; i < iSize; ++i) {
        ipRepetition[i] = 0;
        int j = 0;
        while ((j < i) && (daArray[i] != daArray[j])) {
            if (daArray[i] != daArray[j]) {
                ++j;
            }
        }
        ++(ipRepetition[j]);
    }
    int iMaxRepeat = 0;
    for (int i = 1; i < iSize; ++i) {
        if (ipRepetition[i] > ipRepetition[iMaxRepeat]) {
            iMaxRepeat = i;
        }
    }
    delete [] ipRepetition;
    return daArray[iMaxRepeat];
}


double GetMean(int nums[], int total) {

    int i;
    double sum = 0.0;
    for(i = 0; i < total; i++) {
        sum += nums[i];
    }
    return (sum/total);

}




int main( int argc, char * args[] )
{

    ifstream data_queries("inputfile.txt");      

    string data_querystats = "outputfile.txt";
    ofstream fcoclickout(data_querystats.c_str());

    // Define parameters
    map<string, int> querySet;                                                 // map of queries and their frequencies
    string query;                                                              // hold input query


    // Read query-by-query
    int queryCount = 0;
    while (getline(data_queries, query, '\n'))
    {
        querySet[query]++;                                                     // increase that queries' count
        queryCount++;
    }


    // Print simple statistics
    fcoclickout << "Total number of queries: " << queryCount << endl;
    fcoclickout << "Total number of unique queries: " << querySet.size() << endl;


    // Iterate over map to save counts into array
    int currentMax = 0;
    int idx = 0;
    string maxSugg;
    map<string, int>::iterator map_iterator;
    int Counts[querySet.size()];

    for (map_iterator = querySet.begin(); map_iterator != querySet.end(); map_iterator++)
    {
        if (map_iterator->second > currentMax)
        {
            currentMax = map_iterator->second;
            maxSugg = map_iterator->first;
        }

        // Store counts externally
        Counts[idx] = map_iterator->second;
        idx++;
    }



    // Print advanced statistics
    fcoclickout << "Maximum suggestion count: " << currentMax << endl;
    fcoclickout << "Query with maximum suggestions: " << maxSugg << endl;
    fcoclickout << "Average suggestions/query: " << GetMean(Counts, querySet.size()) << endl;
    fcoclickout << "Median suggestions/query: " << GetMedian(Counts, querySet.size()) << endl;
    fcoclickout << "Mode suggestions/query: " << GetMode(Counts, querySet.size()) << endl;

    cout << "done";
    return 0;

}

[Elysia]

>>double* dpSorted = new double[iSize];
Use vectors. Use the .at() method.

>>map<string, int>::iterator map_iterator;
Prefer to declare these variables inside loops, eg
for (map<string, int>::iterator map_iterator = querySet.begin(); ...)

Or if your compiler supports C++11:
for (auto map_iterator = querySet.begin(), ...)

[jimblumberg]

Also Variable Length Arrays are not allowed in C++:

    int Counts[querySet.size()]; // Not allowed!

I doubt that your median and mean functions are working as desired. Remember a map doesn't allow duplicate keys. Also your input "function" is getting "sentences" not necessarily "words'.




Jim