Convert a C code to Java code of an Artificial Neural Network Problem

//Analysis Type - Regression
#include <stdio.h>
#include <conio.h>
#include <math.h>
#include <stdlib.h>

double input_hidden_weights[4][3]=
{
{5.52706412733837e-001, 1.28925457193726e-001, 5.63971493109330e-001 },
{-1.71178004644907e+000, -1.52596666072101e+000, -5.00996881686563e+000 },
{1.84805441095260e+000, 4.14664704403166e-001, -6.31190329919085e-001 },
{-5.75580725064369e+000, -5.88786472164176e+000, 6.55702460354330e-001 }
};

double hidden_bias[4]={ -1.22139719299858e+000, 2.91772593533840e+000, -9.05733899543194e-001, 3.30978537834433e+000 };

double hidden_output_wts[1][4]=
{
{-2.70667050304236e-001, 8.34208850112333e-001, 8.38407685534142e-002, -8.46927424254966e-001 }
};

double output_bias[1]={ 3.27855408071900e-001 };

double max_input[3]={ 8.00000000000000e+001, 7.90000000000000e+001, 8.40000000000000e+001 };

double min_input[3]={ 2.50000000000000e+001, 2.50000000000000e+001, 2.30000000000000e+001 };

double max_target[1]={ 2.10000000000000e+000 };

double min_target[1]={ 1.00000000000000e+000 };

double input[3];
double hidden[4];
double output[1];

void ScaleInputs(double* input, double minimum, double maximum, int size)
{
double delta;
long i;
for(i=0; i<size; i++)
{
delta = (maximum-minimum)/(max_input[i]-min_input[i]);
input[i] = minimum - delta*min_input[i]+ delta*input[i];
}
}

void UnscaleTargets(double* output, double minimum, double maximum, int size)
{
double delta;
long i;
for(i=0; i<size; i++)
{
delta = (maximum-minimum)/(max_target[i]-min_target[i]);
output[i] = (output[i] - minimum + delta*min_target[i])/delta;
}
}

void ComputeFeedForwardSignals(double* MAT_INOUT,double* V_IN,double* V_OUT, double* V_BIAS,int size1,int size2,int layer)
{
int row,col;
for(row=0;row < size2; row++)
{
V_OUT[row]=0.0;
for(col=0;col<size1;col++)V_OUT[row]+=(*(MAT_INOUT+(row*size1)+col)*V_IN[col]);
V_OUT[row]+=V_BIAS[row];
if(layer==0) V_OUT[row] = tanh(V_OUT[row]);
}
}

void RunNeuralNet_Regression ()
{
ComputeFeedForwardSignals((double*)input_hidden_weights,input,hidden,hidden_bias,3, 4,0);
ComputeFeedForwardSignals((double*)hidden_output_wts,hidden,output,output_bias,4, 1,1);
}

int main()
{
int i=0;
int keyin=1;
while(1)
{
printf("\nEnter values for Continuous inputs\n");
printf("Cont. Input-0(Var1): ");
scanf("%lg",&input[0]);
printf("Cont. Input-1(Var2): ");
scanf("%lg",&input[1]);
printf("Cont. Input-2(Var3): ");
scanf("%lg",&input[2]);
ScaleInputs(input,0,1,3);
RunNeuralNet_Regression();
UnscaleTargets(output,0,1,1);
printf("\nPredicted Output of Var4 = %.14e",output[0]);
printf("\n\nPress any key to make another prediction or enter 0 to quit the program.\n");
keyin=getch();
if(keyin==48)break;
}
return 0;
}