Question

I’ve been experimenting with the FANN library, which seems to be a great library for neural network, and I’m having some issue on how to use it.

So what I’m trying to do here is training a neural network, for the sake of messing with the library, giving it an input and expecting an output.

FANN::neural_net nn;
const float desired_error = 0.00001;
const unsigned int max_epochs = 500000;
const unsigned int epochs_between_reports = 1000;
const unsigned int layers_count = 3;
const unsigned int layers[layers_count] = {7, 5, 1};
nn.create_standard_array(layers_count, layers);
nn.train_on_file(TRAINING_DATA, max_epochs, epochs_between_reports, desired_error);

Here are the first lines of my training data file (TRAINING_DATA) :

16969 7 1
0.0812069 0.0812069 0.381578 0.0812069 5.8931e-05 0.0843302 0.606695 
1
0.429961 0.0509753 0.381578 0.0266957 0.000117862 0.00707172 0.0221581 
1
0.0983558 0.486888 0.381578 0.000117862 0.0266957 0.00701279 0.0539808 
1
0.0983558 0.486888 0.598562 0.0161471 0.0161471 0.000471448 0.00135541 
1

The complete dataset can be found here

Using a sample data from the training data file, I should get the output matching it, right? However, if I do the following, I get 0 as output…

fann_type i[7], *o;
i[0] = 0.429961; i[1] = 0.0509753; i[2] = 0.381578; i[3] = 0.0266957; i[4] = 0.000117862; i[5] = 0.00707172; i[6] = 0.0221581;
o = nn.run(i);
std::cout << "output (run) is " << o[0] << std::endl;

Can someone actually explain me what is going on here?

I use the 2.2.0 version of fann.

Thank you

Edit : It seems that the 2.1.0 beta version give the expected results, but not the 2.2.0 version.

Edit 2 : It was actually a bug in the version I was using.

Answer 1

I tried to reproduce your error but I could not. Here is my program:

#include<iostream>
using namespace std;
#include <fann.h>
#include <fann_cpp.h>
#include <floatfann.h>
int main()
{
  FANN::neural_net nn;
  const float desired_error = 0.00001;
  const unsigned int max_epochs = 500000;
  const unsigned int epochs_between_reports = 1000;
  const unsigned int layers_count = 3;
  const unsigned int layers[layers_count] = {7, 5, 1};
  nn.create_standard_array(layers_count, layers);
  nn.train_on_file("test.train", max_epochs, epochs_between_reports, desired_error);

  fann_type i[7];
  i[0] = 0.429961; i[1] = 0.0509753; i[2] = 0.381578; i[3] = 0.0266957; i[4] = 0.000117862; i[5] = 0.00707172; i[6] = 0.0221581;
  fann_type *o = nn.run(i);
  std::cout << "output (run) is " << o[0] << std::endl;

  return 0;
}

This is the output:

Max epochs   500000. Desired error: 0.0000100000.
Epochs            1. Current error: 0.2283857614. Bit fail 4.
Epochs            7. Current error: 0.0000000000. Bit fail 0.
output (run) is 1

Maybe you could provide your full training set?