Plotting shapes based on equations and weights.

Let us say I have weights w0, w1,w2 and I want to fit a line such that w0(1)+w1x+w2y=0 or any other shape, how do I do it? Is there a specific function that accepts the weights and the degree of variable/axis?

I can accept answers to question with reference to or pertanent to Machine Learning boundaries for classification.

Thanks in advance :)

[Walter Roberson]

What is the input data? What output values are you attempting to find?

It looks to me as if you have the linear equation

y = - w1/w2 * x - w0(1)/w2

and that you might be hoping to fit for unknowns w1, w2, and w0(1) .

If that is the case, then it cannot be done: you can only fit for the ratios w1/w2 and w0(1)/w2

However, if one of those values is already known then you can fit for the other two.

[Anmol Pardeshi]

I am not trying to find these weights. I already have the values of these weights - for example w0=0.5 ; w1=-3 ; w2=-0.5.

What I want to know is ff there is a function available where I can pass these weights and the equations I want it to fit in and the function just outputs the plot. (In this case if I pass the function with my weights and my function, it plots a straight line for me.)

Answer 1

[Matt J]

For 2D and 3D functions, you can use fimplict and fimplicit3, e.g.,

fimplicit(@(x,y) w0+w1*x+w2*y)