Finding a Transformation Matrix for Matching Signs in a Matrix

  • Context: Undergrad 
  • Thread starter Thread starter Phong
  • Start date Start date
  • Tags Tags
    Matrix
Click For Summary
SUMMARY

The discussion centers on calculating a transformation matrix to match the signs of a source matrix to a target matrix. The user, Nhat, attempts to use the inverse of the target matrix multiplied by the source matrix but encounters difficulties. A suggestion is made to simplify the process by directly flipping the signs of relevant entries instead of relying on matrix multiplication. The proposed formula for matrix transformation is T=(TS-1)S.

PREREQUISITES
  • Understanding of matrix operations, specifically matrix multiplication and inversion
  • Familiarity with transformation matrices in linear algebra
  • Knowledge of sign manipulation in matrices
  • Basic programming concepts for implementing matrix operations
NEXT STEPS
  • Research matrix inversion techniques in linear algebra
  • Learn about transformation matrices and their applications
  • Explore sign manipulation methods in matrix computations
  • Investigate programming libraries for matrix operations, such as NumPy in Python
USEFUL FOR

Mathematicians, computer scientists, and software developers working with linear algebra and matrix transformations, particularly those involved in graphics programming or data analysis.

Phong
Messages
5
Reaction score
0
Hi!

I have two sets of matrices where only the sign of each field might differ. I'm trying to calculate a generally applicable matrix that transforms the signs of the source matrix so it has the same signs like the target matrix. I tried calculating the inverse of the target matrix and multiply that with the source, but I'm missing a step. Maybe you guys can help!

Thanks a lot!


Nhat



TARGET
-1.05529 -0.707107 0.707107 0
-6.18172 -0.707107 -0.707107 0
1 -1.18331 -3.09086 0
-5 5.91657 -14.1421 1

SOURCE
1.05529 -0.707107 -0.707107 0
6.18172 -0.707107 0.707107 0
-1 -1.18331 -3.09086 0
-5 -5.91657 -14.1421 1
 
Physics news on Phys.org
Seems odd to use matrix multiplication for this, when you can just flip the sign of the relevant entries. (The numbers make me think that this is for a computer program. It would be easier both for you and the program to work with the components directly). If you insist on using matrix multiplication, you would have to do something like this:
$$T=(TS^{-1})S$$
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 7 ·
Replies
7
Views
4K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 34 ·
2
Replies
34
Views
3K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K