Im fairly new to c++ and i would like to ask for suggestions / advice if there is a better / more optimal way to use a function calling ifstream and string stream.
I have a document with the structure with 150 lines and 8 columns (a small subset with values simplified):
5.43e-08 0.0013 0.0105 0.013 0.026 0.068 0.216 0.663
6.98e-08 0.0004 0.0188 0.022 0.103 0.854 0 0
7.31e-08 0.0004 0.0125 0.017 0.074 0.895 0 0
5.82e-08 0.0006 0.0596 0.075 0.150 0.713 0 0
the number of each line represents a position (pos 1 ... pos 150) and each column is a probability of a quality (Qual1 .. Qual8). My goal is to sample from each line being a quality distribution to create a string of qualities for the 150 positions. I have created a function which can do this.
std::string Qual(std::ifstream &infile){
std::string line;
double Q_1,Q_2,Q_3,Q_4,Q_5,Q_6,Q_7,Q_8;
char Qualities[] = {'1', '2', '3', '4' ,'5', '6', '7','8','\0'};
std::string Read_qual;
while (std::getline(infile, line)){
std::stringstream ss(line);
ss >> Q_1 >> Q_2 >> Q_3 >> Q_4 >> Q_5 >> Q_6 >> Q_7 >> Q_8;
std::srand(std::time(nullptr));
std::random_device rd;
std::default_random_engine gen(rd());
std::discrete_distribution<> d({Q_1,Q_2,Q_3,Q_4,Q_5,Q_6,Q_7,Q_8});
Read_qual += Qualities[d(gen)];
}
return Read_qual;
}
The problem is that I have to use this function repeatedly to create multiple of these distributions based on some other input. And as far as I can read here on stack overflow I have to use .clear() and seekq to keep the file open but still use it.
int main(int argc,char **argv){
std::ifstream infile("Freq.txt");
std::cout << Qual(infile) << std::endl;
infile.clear();
infile.seekg(0);
std::cout << "-------" << std::endl;
std::cout << Qual(infile);
return 0;
}
My question is: Is there a more ideal solution to accomplish this when using c++. Like any functions which are perhaps faster. Could anyone come with any suggestions? is it better to keep opening and closing the file?
Aucun commentaire:
Enregistrer un commentaire