Rank: Advanced Member Groups: Registered
Joined: 28/08/2014(UTC) Posts: 1,356 Was thanked: 815 time(s) in 516 post(s)
|
Hi. Something is wrong here, in both MProd functions, but can't figure what is it. Thanks in advance. mprod bug.sm (13kb) downloaded 71 time(s).Alvaro.
|
|
|
|
Rank: Guest
Groups: Registered
Joined: 04/07/2015(UTC) Posts: 6,866 Was thanked: 981 time(s) in 809 post(s)
|
Originally Posted by: Razonar Something is wrong here, in both MProd functions, but can't figure what is it. Hola, Alvaro. Not sure what you are attempting to do 1. Multiply a two cols matrix by a single number [2D rotation]. 2. Multiply a 3 cols matrix by a 3x3 matrix [3D rotation] 3D rotation from Smath 3D [3 cols] creator. Both exemplified Annalema. 3. RGB nxm matrix rotation [example Venturi RGB] 4. Square matrix [nxn] rotation [example Hyper-Surface] Thanks Alvaro ... Jean Image Rotate Matrix.sm (42kb) downloaded 51 time(s). Analemma.sm (141kb) downloaded 29 time(s). Image Rotate RGB.sm (349kb) downloaded 25 time(s).
|
|
|
|
Rank: Member Groups: Registered
Joined: 28/11/2017(UTC) Posts: 20 Location: Pisa Was thanked: 2 time(s) in 2 post(s)
|
Originally Posted by: Razonar Hi. Something is wrong here, in both MProd functions, but can't figure what is it. Thanks in advance. mprod bug.sm (13kb) downloaded 71 time(s).Alvaro. Hello Alvaro, I can't explain the problem, but I see that it originates from the way arguments are passed to MDot. See attachment, writing arguments directly into MDot gives correct result, passing through row() and col() gives an "a not defined" error. It seems that using row() and/or col() a numerical evaluation is triggered, even if the region is marked as "Symbolic". Sorry, I can't give a better explanation. Marco mprod_bug_MBe.sm (17kb) downloaded 37 time(s).P.S.: for matrix multiplication also index notation can be used, which works pretty nice in Smath. I attach here an example, just in case it could be useful to somebody. I wonder if Einstein notation could be implemented in Smath, so that summation symbols could be implicitly applied on mute indexes... maybe this is aking too much... IndexNotation.sm (12kb) downloaded 41 time(s).Edited by user 02 December 2019 20:05:26(UTC)
| Reason: Not specified
|
1 user thanked MBernacca for this useful post.
|
|
|
Rank: Guest
Groups: Registered
Joined: 04/07/2015(UTC) Posts: 6,866 Was thanked: 981 time(s) in 809 post(s)
|
Originally Posted by: MBernacca Hello Alvaro, I can't explain the problem, but I see that it originates from the way arguments are passed to MDot. See attachment, writing arguments directly into MDot gives correct result, passing through row() and col() gives an "a not defined" error. It seems that using row() and/or col() a numerical evaluation is triggered, even if the region is marked as "Symbolic". Sorry, I can't give a better explanation. BTW: Thanks Marco for the Scilab/Smath Vector Intersection code. I encountered many instances row(,,) col(,,) that must be assigned as an algo My understanding is like this: 1. Assume Alvaro has an 'n col vector A' 2. He wants to multiply any col or all [even in mixed order] 3. Multiplication by a number is immediate. 4. Multiplication by 'L length vector' needs a vectorize algo. In the soon/next posted Rainbow it is ± what we are doing. Instead of multiplying, we linterp. Jean
|
|
|
|
Rank: Advanced Member Groups: Registered
Joined: 28/08/2014(UTC) Posts: 1,356 Was thanked: 815 time(s) in 516 post(s)
|
Originally Posted by: MBernacca Originally Posted by: Razonar Hi. Something is wrong here, in both MProd functions, but can't figure what is it. Thanks in advance. mprod bug.sm (13kb) downloaded 71 time(s).Alvaro. Hello Alvaro, I can't explain the problem, but I see that it originates from the way arguments are passed to MDot. See attachment, writing arguments directly into MDot gives correct result, passing through row() and col() gives an "a not defined" error. It seems that using row() and/or col() a numerical evaluation is triggered, even if the region is marked as "Symbolic". Sorry, I can't give a better explanation. Marco mprod_bug_MBe.sm (17kb) downloaded 37 time(s).P.S.: for matrix multiplication also index notation can be used, which works pretty nice in Smath. I attach here an example, just in case it could be useful to somebody. I wonder if Einstein notation could be implemented in Smath, so that summation symbols could be implicitly applied on mute indexes... maybe this is aking too much... IndexNotation.sm (12kb) downloaded 41 time(s). Hi Marco. Thanks very much for your appointments. I go to take them in consideration for see the best way to extend dot product for some simple manipulations with symbolic algebra and taking the conjugate transpose instead the second argument. Best regards. Alvaro.
|
|
|
|
Rank: Advanced Member Groups: Registered
Joined: 28/08/2014(UTC) Posts: 1,356 Was thanked: 815 time(s) in 516 post(s)
|
Originally Posted by: Jean Giraud Originally Posted by: Razonar Something is wrong here, in both MProd functions, but can't figure what is it. Hola, Alvaro. Not sure what you are attempting to do ... Hi Jean. Just make very simple math, checking how to write the elemental things in SMath. If not, how to write something more complex? Best regards. Alvaro.
|
|
|
|
Rank: Advanced Member Groups: Registered
Joined: 28/08/2014(UTC) Posts: 1,356 Was thanked: 815 time(s) in 516 post(s)
|
Originally Posted by: MBernacca It seems that using row() and/or col() a numerical evaluation is triggered, even if the region is marked as "Symbolic". Sorry, I can't give a better explanation.
Marco
Hi. Enogh to know that problem it's in row and col functions. Even I can't write the code for them, seems to be solved the issue rewriting length function. mprod_workaround.sm (8kb) downloaded 25 time(s).Same solution it's used for Lagrange equations with vectors. Best regards. Alvaro.
|
|
|
|
Rank: Guest
Groups: Registered
Joined: 04/07/2015(UTC) Posts: 6,866 Was thanked: 981 time(s) in 809 post(s)
|
Originally Posted by: Razonar Even I can't write the code for them, Alvaro ... is that what you are looking for ? Utilities Matrix Product[X,Y].sm (17kb) downloaded 25 time(s).
|
|
|
|
Rank: Member Groups: Registered
Joined: 28/11/2017(UTC) Posts: 20 Location: Pisa Was thanked: 2 time(s) in 2 post(s)
|
Originally Posted by: Razonar Originally Posted by: MBernacca It seems that using row() and/or col() a numerical evaluation is triggered, even if the region is marked as "Symbolic". Sorry, I can't give a better explanation.
Marco
Hi. Enogh to know that problem it's in row and col functions. Even I can't write the code for them, seems to be solved the issue rewriting length function. mprod_workaround.sm (8kb) downloaded 25 time(s).Same solution it's used for Lagrange equations with vectors. Best regards. Alvaro. Hello everybody. More on the strange behaviour of function "lenght()". See attached. Anybody can explain why lenght() does not accept the evaluation operator? Thanks Marco Lenght_MBe.sm (9kb) downloaded 17 time(s).P.S.: Sorry, my mistake. Do not consider this post. I just mispelled the function... Edited by user 04 December 2019 18:34:03(UTC)
| Reason: Not specified
|
|
|
|
Rank: Guest
Groups: Registered
Joined: 04/07/2015(UTC) Posts: 6,866 Was thanked: 981 time(s) in 809 post(s)
|
Originally Posted by: MBernacca Anybody can explain why lenght() does not accept the evaluation operator? Maybe I don't understand ...
|
|
|
|
Forum Jump
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.