firstann.cpp

#include <vector>
#include <fstream>
#include <iostream>
#include <ctype.h>
#include "fann.h"
#include "fann_cpp.h"

using namespace std;

void error(const char* p, const char* p2 = ""){
    cerr << p << ' ' << p2 << endl;
    exit(1);
}

/**
    Generates letter frequencies from a file to add to the training data
*/
void generate_frequencies(const char *filename, float *frequencies){
    ifstream infile(filename);
    if(!infile) error("Cannot open input file", filename);
    vector<unsigned int> letter_count(26, 0);
    unsigned int num_characters = 0;
    char c;
    while(infile.get(c)){
        c = tolower(c);
        if(c >= 'a' && c <= 'z'){
            letter_count[c - 'a']++;
            num_characters++;
        }
    }
    if(!infile.eof()) error("Something strange happened");
    for(unsigned int i = 0; i != 26; i++) frequencies[i] = letter_count[i]/(double)num_characters;
}

/**
    Generates letter frequencies from a file to add to the training data
*/
/*int main(int argc, char* argv[]){
    if(argc != 2) error("Remember to specify an input file");

    float frequencies[26];
    generate_frequencies(argv[1], frequencies);
    for(unsigned int i = 0; i != 26; i++) cout << frequencies[i] << ' ';

    cout << endl;
    return 0;
}*/

/**
    Tests a file and tells where it's english, french, or polish
*/
int main(int argc, char* argv[]){
    if(argc != 2) error("Remember to specify an input file");

    float frequencies[26];
    generate_frequencies(argv[1], frequencies);
    for(unsigned int i = 0; i != 26; i++) cout << frequencies[i] << ' ';

    FANN::neural_net ann;
    ann.create_from_file("language_classify.net");

    float *output = ann.run(frequencies);

    cout << "English: " << output[0] << endl
        << "French: " << output[1] << endl
        << "Polish: " << output[2] << endl;
    return 0;
}

/**
    Creates the neural network and trains it on the data file
*/
/*int main(){

    const unsigned int num_layers = 3;
    const unsigned int num_input = 26;
    const unsigned int num_hidden = 13;
    const unsigned int num_output = 3;

    const unsigned int layers[3] = {26,26,3};

    FANN::neural_net ann;
    ann.create_standard_array(3, layers);

    ann.train_on_file("frequencies.data", 2000, 10, 0.0001);
    ann.save("language_classify.net");
    ann.destroy();
    return 0;
}*/

training-data.txt

12 26 3
0.103 0.016 0.054 0.060 0.113 0.010 0.010 0.048 0.056 0.003 0.010 0.035 0.014 0.065 0.075 0.013 0.000 0.051 0.083 0.111 0.030 0.008 0.019 0.000 0.016 0.000
1 0 0
0.076 0.010 0.022 0.039 0.151 0.013 0.009 0.009 0.081 0.001 0.000 0.058 0.024 0.074 0.061 0.030 0.011 0.069 0.100 0.074 0.059 0.015 0.000 0.009 0.003 0.003
0 1 0
0.088 0.016 0.030 0.034 0.089 0.004 0.011 0.023 0.071 0.032 0.030 0.025 0.047 0.058 0.093 0.040 0.000 0.062 0.044 0.035 0.039 0.002 0.044 0.000 0.037 0.046
0 0 1
0.0953608 0.0206186 0.0289233 0.0441008 0.119989 0.0214777 0.0177549 0.0395189 0.0693013 0.00171821 0.00515464 0.0395189 0.0252005 0.064433 0.0827606 0.0237686 0.000286369 0.0630011 0.0715922 0.111111 0.0237686 0.00715922 0.00973654 0.00200458 0.0111684 0.000572738
1 0 0
0.0812233 0.00559027 0.0335416 0.0345281 0.14436 0.016442 0.020388 0.00887866 0.0825386 0.000986518 0 0.0723446 0.0272937 0.0730023 0.0473528 0.0437356 0.00427491 0.0792502 0.083854 0.0670832 0.057218 0.0092075 0.000328839 0.00493259 0.00131536 0.000328839 
0 1 0
0.102384 0.0140252 0.0357644 0.0392707 0.0806452 0.00420757 0.0154278 0.013324 0.0841515 0.0189341 0.0518934 0.0399719 0.0273492 0.0631136 0.0806452 0.0224404 0.000701262 0.0350631 0.0575035 0.056101 0.0329593 0.00210379 0.0399719 0 0.0336606 0.0483871 
0 0 1
0.0754835 0.0166348 0.0366944 0.036144 0.129195 0.0204419 0.0168183 0.0430242 0.0943964 0.00241572 0.00455623 0.0426573 0.0250745 0.0796575 0.0641235 0.02363881 0.0139133 0.00180414 0.0158245 0.00223225 
1 0 0
0.0728756 0.0115966 0.0347657 0.0388076 0.151838 0.011789 0.0109951 0.0140266 0.0912809 0.00317583 0.00228563 0.0582235 0.0286546 0.0822106 0.0524974 0.028799 0.0109229 0.0690501 0.0762439 0.0720094 0.0510779 0.0142672 0.00202098 0.00437879 0.00481186 0.00139544
0 1 0
0.102671 0.0212689 0.0358034 0.0299306 0.0970478 0.00689311 0.0164618 0.0131287 0.0970931 0.0206567 0.0262346 0.0265294 0.0265521 0.076255 0.0681375 0.0268242 6.80241e-05 0.0551449 0.0503152 0.0538071 0.0221532 0.00108839 0.0447599 0.000453494 0.0375947 0.0431273 
0 0 1
0.0787958 0.0158488 0.0399622 0.0349005 0.126359 0.0174751 0.0185539 0.0327929 0.0825298 0.00137743 0.00685398 0.0399788 0.0258891 0.0829115 0.0741822 0.0247274 0.000796588 0.0700665 0.0666645 0.0889689 0.0252419 0.0125629 0.015185 0.00217402 0.0135254 0.00167615 
1 0 0
0.0701951 0.0102924 0.0399208 0.0411266 0.142759 0.0143405 0.0110676 0.00701951 0.0817794 0.00198097 0.00189484 0.0520219 0.0285087 0.086172 0.0619698 0.0310925 0.00736402 0.0776452 0.0756212 0.0837604 0.0523233 0.0101201 0.00284225 0.00430645 0.00335903 0.000516774 
0 1 0
0.0715975 0.0082006 0.0348525 0.0329601 0.0928876 0.00583504 0.0138779 0.0102507 0.0815329 0.0205015 0.0329601 0.0206592 0.0247595 0.0769595 0.0851601 0.0392683 0.000157704 0.0608737 0.0492036 0.0756978 0.0332755 0.00141933 0.0517269 0.000946223 0.0370604 0.0373758
0 0 1