How we can use NonlinearLeastSquares ?

[anism...@gmail.com]

Hi Cezar,

Can you please provide us an example to know how we can use NonlinearLeastSquares class ?

Thank you
Best Regards.

[César]

Hi there,

Sure, here it is. Hope it helps!

Best regards,

Cesar

// Suppose we would like to map the continuous values in the
// second column to the integer values in the first column.
double[,] data =
{
    { -40,    -21142.1111111111 },
    { -30,    -21330.1111111111 },
    { -20,    -12036.1111111111 },
    { -10,      7255.3888888889 },
    {   0,     32474.8888888889 },
    {  10,     32474.8888888889 },
    {  20,      9060.8888888889 },
    {  30,    -11628.1111111111 },
    {  40,    -15129.6111111111 },
};


// Extract inputs and outputs
double[][] inputs = data.GetColumn(0).ToArray();
double[] outputs = data.GetColumn(1);


// Create a Nonlinear regression using
var regression = new NonlinearRegression(3,


    // Let's assume a quadratic model function: ax² + bx + c
    function: (w, x) => w[0] * x[0] * x[0] + w[1] * x[0] + w[2],


    // Derivative in respect to the weights:
    gradient: (w, x, r) =>
    {
        r[0] = 2 * w[0]; // w.r.t a: 2a  
        r[1] = w[1];     // w.r.t b: b
        r[2] = w[2];     // w.r.t c: 0
    }
);


// Create a non-linear least squares teacher
var nls = new NonlinearLeastSquares(regression);


// Initialize to some random values
regression.Coefficients[0] = 4.2;
regression.Coefficients[1] = 0.3;
regression.Coefficients[2] = 1;


// Run the function estimation algorithm
double error;
for (int i = 0; i < 100; i++)
    error = nls.Run(inputs, outputs);


// Use the function to compute the input values
double[] predict = inputs.Apply(regression.Compute);

[anism...@gmail.com]

Thank You Cesar it of course help me.
Thank you again.