Background Linear regression is the process of fitting a set

Background Linear regression is the process of fitting a set of data pairs to a straight line. The assumption is that the data can be modeled by a linear relationship. Our task will be to extract from the seemingly noise-corrupted data the parameters of the linear model. We are given a set of data pairs, x.y, osi

Solution

#include #include #define MAX_NUMBER_OF_LINES 10 // parse a maximum of 10 lines #define MAX_LINE_SIZE 50 // parse a maximum of 50 chars per line int main () { FILE* fh = fopen(\"file.txt\", \"r\"); // open the file char temp[MAX_LINE_SIZE]; // some buffer storage for each line // storage for MAX_NUMBER_OF_LINES integers int d_out[MAX_NUMBER_OF_LINES]; // storage for MAX_NUMBER_OF_LINES strings each MAX_LINE_SIZE chars long char s_out[MAX_NUMBER_OF_LINES][MAX_LINE_SIZE]; // i is a special variable that tells us if we\'re parsing a number or a string (0 for num, 1 for string) // di and si are indices to keep track of which line we\'re currently handling int i = 0, di = 0, si = 0; while (fgets(temp, MAX_LINE_SIZE, fh) && di < MAX_NUMBER_OF_LINES) // read the input file and parse the string { temp[strlen(temp) -1] = \'\\0\'; // get rid of the newline in the buffer char* c = strtok(temp, \" \"); // set the delimiters while(c != NULL) { if (i == 0) // i equal to 0 means we\'re parsing a number { i = 1; // next we\'ll parse a string, let\'s indicate that sscanf(c, \"%d\", &d_out[di++]); } else // i must be 1 parsing a string { i = 0; // next we\'ll parse a number sprintf(s_out[si++], \"%s\", c); } c = strtok(NULL, \" \"); } printf(\"%d %s\ \", d_out[di -1], s_out[si - 1]); // print what we\'ve extracted } fclose(fh); return 0; }