MLP

#include <opencv2/core.hpp>
#include <opencv2/imgcodecs.hpp>
#include <opencv2/highgui.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#include <opencv2/ml/ml.hpp>

#include <iostream>
#include <iomanip>

using namespace cv;
using namespace ml;
using namespace std;

#include "diaconverter.hpp"

void print(Mat& mat, int prec)
{
    for (int i = 0; i<mat.size().height; i++)
    {
        cout << "[";
        for (int j = 0; j<mat.size().width; j++)
        {
            cout << fixed << setw(2) << setprecision(prec) << mat.at<float>(i, j);
            if (j != mat.size().width - 1)
                cout << ", ";
            else
                cout << "]" << endl;
        }
    }
}

int main()
{
    const int hiddenLayerSize = 10;
	int trainSamples = 512;

    float inputTrainingDataArray[trainSamples][1];
    float outputTrainingDataArray[trainSamples][10];

	for(int i=0;i<trainSamples;i++)
	{
		int r = (rand() % 1023);
		inputTrainingDataArray[i][0] = r;

		vector<int> a = int2bin(r);

		for(int ant=0;ant<a.size();ant++)
		{
			outputTrainingDataArray[i][ant] = a[ant];
		}

		bool test = true;

		for(int x=0;x<a.size();x++)
		{
			if(a[x] != 0 && a[x] != 1) test = false;
		}

		if(!test) i--;
	}


    Mat inputTrainingData = Mat(trainSamples, 1, CV_32F, inputTrainingDataArray);

    Mat outputTrainingData = Mat(trainSamples, 10, CV_32F, outputTrainingDataArray);

    Ptr<ANN_MLP> mlp = ANN_MLP::create();

    Mat layersSize = Mat(5, 1, CV_16U);
    layersSize.row(0) = Scalar(inputTrainingData.cols);
    layersSize.row(1) = Scalar(hiddenLayerSize);
    layersSize.row(2) = Scalar(hiddenLayerSize);
    layersSize.row(3) = Scalar(hiddenLayerSize);
    layersSize.row(4) = Scalar(outputTrainingData.cols);
    mlp->setLayerSizes(layersSize);

    mlp->setActivationFunction(ml::ANN_MLP::SIGMOID_SYM,0,0);

    TermCriteria termCrit = TermCriteria(TermCriteria::MAX_ITER+TermCriteria::EPS, 900, 0.000000001);

    mlp->setTermCriteria(termCrit);

    mlp->setTrainMethod(ml::ANN_MLP::BACKPROP, 0.000000001);

    Ptr<TrainData> trainingData = TrainData::create(
        inputTrainingData,
        SampleTypes::ROW_SAMPLE,
        outputTrainingData
    );

    mlp->train(trainingData, ANN_MLP::TrainFlags::UPDATE_WEIGHTS | ANN_MLP::TrainFlags::NO_INPUT_SCALE | ANN_MLP::TrainFlags::NO_OUTPUT_SCALE);

	Mat sample = Mat(1, inputTrainingData.cols, CV_32F, 128);
	Mat result;
	mlp->predict(sample, result);
	cout << sample << " -> ";// << result << endl;
	print(result, 0);
	cout << endl;

    return 0;
}