Beantwortet
Eigenvalue doesn't zero the characteristic polynomial
The reason for this is that the characteristic polynomial for matrices of even moderate size tends to be ill-conditioned, no mat...

mehr als 7 Jahre vor | 2

| akzeptiert

Beantwortet
eigs does not return the eigenvalues closest to shift sigma
Hi Olivier, Yes, whether it's preferable for performance to compute more eigenvalues than needed, or to factorize the matrix ...

mehr als 7 Jahre vor | 0

mehr als 7 Jahre vor | 0

Beantwortet
chol() gives error for a (barely) positive definite matrix.
Short answer first: CHOL requires the input matrix to be positive definite, it does not support positive semi-definite. I'll exp...

mehr als 7 Jahre vor | 7

Beantwortet
eigs does not return the eigenvalues closest to shift sigma
I have not heard of this particular issue before, but I could see it happening when the eigenvalues are very clustered. If yo...

mehr als 7 Jahre vor | 0

Beantwortet
How to create a triangulation from a list of edges and list of nodes?
The problem is that a graph is a more general data structure than a Delaunay triangulation. Many graphs do not represent a trian...

mehr als 7 Jahre vor | 0

Beantwortet
optimize sequence of matlab operations for solving a system of linear equations
I assume that most of the time is spent in the line |R = K\E| (as this has cubic complexity in the size of A, while the other op...

mehr als 7 Jahre vor | 0

Beantwortet
relation between principal eigenvector and principal generalized eigenvector
What may be causing the differences you see is that eigs(A, B, k) first checks if the matrix B is symmetric positive definite. I...

mehr als 7 Jahre vor | 0

Beantwortet
Is there any way to reduce the error in the calculation of eigenvectors when using eig function?
I'm just catching up on MATLAB Answers: Can you tell us some more about your problem? How are you measuring the error in the eig...

mehr als 7 Jahre vor | 0

Beantwortet
it is possible to get 30% of eigenvalues in short times by using the eigs(A,number, scalar) function? (A is sparse Hermitian matrix (100,000×100,000))
It's going to be expensive, but I think calling eigs(A, number, scalar) repeatedly is the most realistic way to compute 30% of t...

mehr als 7 Jahre vor | 0

Beantwortet
Matlab flips the eigenvalue and eigenvector of matrix when passing through singularity
The eigenvalues don't have any intrinsic order, so for real symmetric matrices, EIG just sorts them by size. Looking at just the...

mehr als 7 Jahre vor | 1

| akzeptiert

Beantwortet
When is eigs profitable in terms of time comparing to eig?
There is no clear answer, since this depends on many factors: - Are your matrices dense or sparse, and if they are sparse, how ...

mehr als 7 Jahre vor | 0

| akzeptiert

Beantwortet
Right and Left Eigenvectors
The matrices V and W contain all eigenvectors of A. For example |V(:, 1)| and |D(1, 1)| are the right eigenvector and eigenva...

mehr als 7 Jahre vor | 0

Beantwortet
How can I find circle diameter using arc ?
There's a recent blog post by Loren on fitting a sine wave to a picture by eye: <https://blogs.mathworks.com/loren/2018/06/04...

mehr als 7 Jahre vor | 0

Beantwortet
Calculating gradient for equation in higher dimension?
If the matrices and ND-arrays are of a fixed dimension, you can create symbolic variables for them: A = sym('A', [3 3 3]) ...

mehr als 7 Jahre vor | 1

Beantwortet
Matlab internal parallelization not working (?) under linux
You can also use maxNumCompThreads to set the maximum number of computational threads you want MATLAB to use. So for example, yo...

mehr als 7 Jahre vor | 0

Beantwortet
Hellow, I have a problem To obtain participation factors, and specifically with the right and left eigenvectors in a paper I was reading about it, related in Power sistems, so Im tryin to use eig but the result are not the same: thank you 4uhelp
Hi, The eigenvalues and right-side eigenvectors are correct, they are just returned in a different order than you assume: ...

mehr als 7 Jahre vor | 2

| akzeptiert

Beantwortet
Using sparse/full to solve Ax = b
If A is sparse and b is dense, x = A\b returns a dense matrix x, so no need for the "full" command. It's also not necessary to m...

fast 8 Jahre vor | 1

Beantwortet
Decomposition object saveobj method not supported?
This is a limitation on the decomposition object: It cannot be saved into .mat file. The background for this is that some types ...

fast 8 Jahre vor | 1

| akzeptiert

Beantwortet
Matlab internal parallelization not working (?) under linux
You can use the function |maxNumCompThreads| to see the maximum amount of threads MATLAB uses.

fast 8 Jahre vor | 0

Beantwortet
bugs in eig?
Both answers are correct. The eigenvectors of a matrix are not uniquely determined: Every eigenvector can be multiplied by an ar...

fast 8 Jahre vor | 1

Beantwortet
How to identify repeated eigenvalues of a matrix?
For general matrices, the eigenvalues will typically have a bit of round-off error, so repeated eigenvalues will not be exactly ...

fast 8 Jahre vor | 0

Beantwortet
How to add edges to a graph plot using GraphPlot handle?
There is currently no way to do this, the edges and nodes must be modified on the |graph| or |digraph|, and then this can be pas...

fast 8 Jahre vor | 0

| akzeptiert

Beantwortet
Eigs yields different results on different versions
Numerical results can change between versions, and for eigenvectors, it's possible that several eigenvectors have the same or ve...

fast 8 Jahre vor | 0

Beantwortet
How to improve speed of calculating trace in a script?
You can make the trace operator work faster as follows: Currently, the input is two truncated SVDs, |A1 = U1 * S1 * V1'| and |A2...

fast 8 Jahre vor | 1

| akzeptiert

Beantwortet
How to find the permutation matrix for the qr decomposition?
You need to compute all three matrices together: Q and R are different if the third output e is also returned. For an algorit...

fast 8 Jahre vor | 2

| akzeptiert

Beantwortet
How to find the permutation matrix for the qr decomposition?
I'm afraid using the three-output syntax is the only way. You can use [~, ~, p] = qr(A); but Q and R are still computed,...

fast 8 Jahre vor | 0

Beantwortet
Nonconvex: Checking positive-semidefiniteness inside the Gloptipoly package
For positive definiteness, I would recommended to check this using CHOL (see if the second output registers a problem), but this...

fast 8 Jahre vor | 0

| akzeptiert

Beantwortet
Order of eigenvalues depending on matrix equation order of operations
The reason for the difference in this case is that matrix B is exactly symmetric, and matrix A is not. EIG detects if the input ...

fast 8 Jahre vor | 1

| akzeptiert

Beantwortet
Examining entries in binary matrices and creating new group matrix based on % of connections (defined by 1)
If you have each matrix stored in a separate variable, you can do Atotal = A1 + A2 + A3; % Sum of logical matrices is...

fast 8 Jahre vor | 0