Answered
Inconsistency in behavior of SVD
Firstly, I agree completely with John's great explanations above: Any singular value below eps*first singular value should be tr...

3 years ago | 3

| accepted

Answered
Accuracy Problem when solving linear equation system using: lu(S) vs. decomposition(S,'lu')
The version in decomposition does some optional steps of iterative refinement: It uses the same solution you have above based on...

3 years ago | 0

| accepted

Answered
Need help fixing errors on code written in old MATLAB syntax
I'd also recommend replacing the calls svd(A) and svd(A, 0) with calling svd(A, 'econ'), since the code here is only using the f...

3 years ago | 0

Answered
Different results on different computers, Matlab 2021b - 64 bit, windows 64 bit, both Intel chips
First, about "which value to trust?" Both values are equally trustworthy, the differences in results come down to applying the ...

3 years ago | 1

| accepted

Answered
eigs with Extended Capabilities
The first of these would only be useful if you need to apply EIGS to many problems in parallel, in which case each of these coul...

3 years ago | 1

Answered
How to parallelize many small SVD's on a GPU
As of R2021b, there is a function pagesvd which is supported on both CPU and distributed arrays. It's not supported for gpuArray...

3 years ago | 2

| accepted

Answered
How does shortestpath function work?
Every edge has a number, which is the order in which they appear in the Edges table (try displaying g.Edges). The edgepath conta...

3 years ago | 1

| accepted

Answered
Eigs passes the wrong StartVector
The standard method used in EIGS is only efficient for matrices that are quite large. When eigs detects that the input matrix is...

3 years ago | 0

| accepted

Answered
How to enhance the performance of Matalb's function sqrtm?
Not in general. You might be able to find a cheaper but less accurate way to compute this, but if that's acceptable would depend...

3 years ago | 0

Answered
fastest path between more than two nodes
To compute simply the shortest-path distance, you can use the distances function and pass in a graph object you've constructed u...

3 years ago | 1

Answered
Follow a complex eigenvalue in a spectrum at different times
You could try using the matchpairs function: rng default; d = randn(20, 1) + 1i*randn(20, 1); d2 = d + 0.1*(randn(20, 1) + 1i...

3 years ago | 1

| accepted

Answered
Error using eig in Matlab 2021b by using Parallels 17 of Macbook Pro (M1 )
Since MATLAB is trying to load a .dll file, I'm assuming you're running a Windows VM on this Macbook Pro. MATLAB doesn't support...

3 years ago | 0

| accepted

Answered
use of svds with function handles
The svds function needs to be able to apply both A*x and A'*x, so your function handle should accept a second input which is eit...

3 years ago | 2

Answered
How to compute cholesky to all slice of a tensor?
There isn't another way to do this right now. We have functions that do other linear algebra operations to each page of an ND-ar...

3 years ago | 1

Answered
Why does mldivide solve equations so fast that FORTRAN can't compare it?
MATLAB's mldivide uses multithreading. It doesn't use the GPU unless you have the Parallel Computing Toolbox and are using the g...

3 years ago | 0

Answered
How to highlight single edges in an undirected graph?
Do you still want the multiple edges to be displayed? If not, you can use the simplify command to reduce all multiple edges to j...

3 years ago | 1

| accepted

Answered
Unable to find out left eigen vectors of symbolic matrix
The left eigenvectors are still expressed as right eigenvectors of M', meaning they satisfy a slightly different equation: syms...

3 years ago | 0

| accepted

Answered
Program to find connected and non-isomorphic graphs
In the last loop, you are checking if adj is isomorphic to any previous graph. If there is one graph that adj is not isomorphic ...

3 years ago | 0

| accepted

Answered
How can I invert a big matrix e.g. 1000-by-1000
When running inv on this matrix, there's a warning: >> M = inv(current_Fit_Mat); Warning: Matrix is close to singular or badly...

3 years ago | 0

Answered
Add an automatic TOC to an .mlx
You can go the "Insert" tab (just to the right of where it says "Live Editor" on top of the MATLAB window) and there's a button ...

3 years ago | 0

Answered
How to get a Table Contents in Live Script .mlx export to HTLM
You can go the "Insert" tab (just to the right of where it says "Live Editor" on top of the MATLAB window) and there's a button ...

3 years ago | 2

Answered
How is SVDS execution time related to the target rank?
So the goal of svds is usually to compute a requested rank that is much smaller than either size of the input matrix. A search s...

3 years ago | 0

| accepted

Answered
How to solve eigs/checkInputs issue in EOF analysis?
The error here is happening because R eof>mycaleof is an empty matrix, and eof is trying to compute an eigenvalue of it. Looking...

3 years ago | 0

Answered
How to find out isolated subgraph?
The component output in your code is the number of components. You can use the second output of conncomp to get a vector contai...

3 years ago | 1

Answered
Find all cycles (or the initial node of each cycle) in a very large directed graph (12m nodes, 6.7m edges)
Have you tried using 'MaxNumCycles' to only compute the first 100 / 1000 / 1e4 cycles? That could give some indication if it's c...

3 years ago | 1

| accepted

Answered
Finding number of independent pathways in a graph
I'd expect the call to rmedge is the most expensive here. You could instead add edge weights for every edge in the graph, and th...

3 years ago | 0

Answered
Feature for eigs involving just positive eigenvalues
There's no option to do so directly, since it's not a common request and there's already a lot of option names in EIGS to dig th...

3 years ago | 0

| accepted

Answered
Large sparse matrix LU decomposition
If the call to lu runs out of memory, but the call to decomposition(__, 'lu') doesn't, likely the reason is that decomposition u...

3 years ago | 0

| accepted

Answered
Facing a unbalanced matrix, such as 3*5000, why someone reduce the computational cost by svd(A*A') .
The computational complexity of svd (when using the 'econ' option, which is very necessary for matrices that are far from square...

3 years ago | 1

| accepted

Load more