Finding Jacobian matrix for Newton's method
20 次查看(过去 30 天)
显示 更早的评论
I have a very basic newton's method that uses a loop and:
y = Jac(x)\(-F(x));
x = x + y;
to solve for the approximate solution.
Where x is a the initial guess in the form of a vector, F is the nonlinear function, and Jac is the jacobian matrix. Currently, I am inputting the jacobian by hand.
For example, system of equations =
2x(1) + x(2)
3x(1) + x(2)^2
=> Jac(x) =
[2, 1; 3, 2x(2)]
I was wondering if instead of solving it by hand if I could get Matlab to do it for me.
0 个评论
采纳的回答
Walter Roberson
2012-4-13
If you have the symbolic toolbox you can use the jacobian() function.
2 个评论
Walter Roberson
2019-8-8
x = sym('x', [1 2]);
eqn = [2*x(1) + x(2)
3*x(1) + x(2)^2];
jacobian(eqn, x)
更多回答(1 个)
DIPANKAR POREY
2019-8-7
2x(1) + x(2)
3x(1) + x(2)^2
=> Jac(x) =
[2, 1; 3, 2x(2)]
1 个评论
Walter Roberson
2019-8-8
This does not appear to be an answer? It appears to be a copy of part of the question.
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 Operating on Diagonal Matrices 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!