analyze.cpp

   1 /*
   2  * Copyright (c) 2009 Mauro Iazzi
   3  *
   4  * Permission is hereby granted, free of charge, to any person
   5  * obtaining a copy of this software and associated documentation
   6  * files (the "Software"), to deal in the Software without
   7  * restriction, including without limitation the rights to use,
   8  * copy, modify, merge, publish, distribute, sublicense, and/or sell
   9  * copies of the Software, and to permit persons to whom the
  10  * Software is furnished to do so, subject to the following
  11  * conditions:
  12  *
  13  * The above copyright notice and this permission notice shall be
  14  * included in all copies or substantial portions of the Software.
  15  *
  16  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
  17  * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
  18  * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
  19  * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
  20  * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
  21  * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
  22  * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  23  * OTHER DEALINGS IN THE SOFTWARE.
  24  */
  25
  26 #include <string>
  27 #include <vector>
  28 #include <iostream>
  29 #include <fstream>
  30 #include <sstream>
  31 #include <cmath>
  32 #include <cstdlib>
  33
  34 using namespace std;
  35
  36 std::pair<double, double> compute_stats (const std::vector<double> &d) {
  37     double avg = 0.0, var = 0.0;
  38     const int N = d.size();
  39     for (int i=0;i<N;i++) {
  40         //cerr << d[i] << " ";
  41         avg += d[i];
  42         var += d[i]*d[i];
  43     }
  44     avg /= N;
  45     var /= (N);
  46     //cerr.precision(10);
  47     //cerr << d[0] << " " << d[49] << " " << d[311] << " " << d[N-1] << endl;
  48     //cerr << avg*avg << " " << var << endl;
  49     var -= avg*avg;
  50     var = sqrt(var);
  51     //cerr << "=> " << avg << " +- " << var << endl;
  52     return std::pair<double, double>(avg, var);
  53 }
  54
  55 std::string analyze_data_file (std::string filename, const int skip, const int len, int w, bool sq = false) {
  56     std::vector<double> sums;
  57     std::string ret = "";
  58
  59     int n = -skip-1;
  60     double norm = 0.0;
  61     std::string line_string = "";
  62     std::string state = "";
  63     std::ifstream f(filename.c_str(), std::fstream::in);
  64     while (!f.eof()) {
  65         double energy = 0.0;
  66         double weight = 0.0;
  67         getline(f, line_string);
  68         std::stringstream line(line_string);
  69         if (line_string[0]=='#') {
  70             if (line_string[1]=='!') {
  71                 if (line_string.substr(2, 8)=="prepend ") {
  72                     ret.insert(0, " ");
  73                     ret.insert(0, line_string.substr(10, line_string.size()));
  74                 } else {
  75                 }
  76             }
  77         } else {
  78             n++;
  79             if (n>=0) {
  80                 line >> state >> energy;
  81                 if (sq) energy = energy*energy;
  82                 weight = 1.0;
  83                 for (int j=0;j<w;j++) {
  84                     line >> weight;
  85                 }
  86                 norm += (weight);
  87                 if (n%len==0) {
  88                     if (n>0) sums[sums.size()-1] /= norm;
  89                     sums.push_back(0.0);
  90                     norm = 0.0;
  91                 }
  92                 sums[n/len] += energy*(weight);
  93             }
  94         }
  95     }
  96     f.close();
  97     sums.pop_back(); // last one may be incomplete.
  98     // time to do some thing with the sums (now they are averages)
  99     std::pair<double, double> s = compute_stats(sums);
 100     std::stringstream rs;
 101     rs << s.first << " " << s.second;
 102     ret.append(rs.str());
 103     return ret;
 104 }
 105
 106 #define NEXTOPT(a, n) ( (a[i][2]=='\0')?(a[++i]):&(a[i][2]) )
 107
 108 int main (int argc, char **argv) {
 109     int skip = 0;
 110     int len = 1000;
 111     int w_col = -1;
 112     bool square = false;
 113     vector<string> files;
 114     for (int i=1;i<argc;i++) {
 115         if (argv[i][0]=='-') {
 116             switch (argv[i][1]) {
 117                 case 'w':
 118                     w_col = atoi(NEXTOPT(argv, i));
 119                     break;
 120                 case 's':
 121                     skip = atoi(NEXTOPT(argv, i));
 122                     break;
 123                 case 'q':
 124                     square = true;
 125                     break;
 126                 case 'l':
 127                     len = atoi(NEXTOPT(argv, i));
 128                     break;
 129             }
 130         } else {
 131             files.push_back(argv[i]);
 132         }
 133     }
 134     for (vector<string>::iterator iter = files.begin();iter != files.end();iter++) {
 135         cout << analyze_data_file(*iter, skip, len, w_col, square) << endl;
 136     }
 137     return 0;
 138 }
 139