Login [Register]
Don't have an account? Register now to chat, post, use our tools, and much more.
KermMartian wrote:
Matlab's powerful, but it's an amazingly slow scripting language for many things due to the overhead of its flexibility.


Which is why I just think Rayden sucks at Python. He probably wasn't using things like the array module or the numpy module which exist exactly for problems like this Rolling Eyes

And hey, look at this. Matlab is roughly the same speed as Python + numpy (with the nod going to python as things like Psycho will push it past matlib in terms of speed, and it can be optimized to greater effect) http://lbolla.wordpress.com/2007/04/11/numerical-computing-matlab-vs-pythonnumpyweave/
That pretty much narrows it down to rayden sucking at python Rolling Eyes
That's because noobs like you do not know what is vectorization when people do not avoid vectorizing MATLAB IS FASTER THAN PYTHON duh... If those noobs known what is vectorization they would know that MATLAB do not need for loops to calculate the same things that python requires, using loops...
In fact the link you provided uses a function that uses loops

Quote:

The example we will consider is a very simple (read, trivial) case of solving the 2D Laplace equation using an iterative finite difference scheme (four point averaging, Gauss-Seidel or Gauss-Jordan). The formal specification of the problem is as follows. We are required to solve for some unknown function u(x,y) such that \nabla^2 u = 0 with a boundary condition specified. For convenience the domain of interest is considered to be a rectangle and the boundary values at the sides of this rectangle are given.


know when you understand vectorization come to talk with me if you don't understand it do not even try it!!!
I'm sure Kllrnohj understands vectorization and its positive effects on execution speed in Matlab. The only tradeoff is that you tend to use substantially more memory, but even so, the evidence that python is at least as fast as Matlab seems pretty convincing to me...
you aren't seeing what is vectorization as vectorization will not use more memory, in fact when you do a loop you will use mutch more memory!!!

The main fact was that kll have said that i failed because of using an script language for testing my algorythm in fact he took many hours to reply that
Quote:
maybe because he have been googling till he found numpy
, my main objective when i have gone trought python was to grab a language that can execute on any machine without having to load the main program as matlab done in the old days, but python have made me some constrainments has there are some functions that i need that it didn't have like CAS support was missing"do not try to go over me because i know that it now haves SYMPY for that".

Now Matlab has a feature called mcc that permits to compile MatLab programs, thus not needing the MatLab program installed in a computer to execute that program... If you do not trust me go google and search for MATLAB mcc command, this tiny command as great inputs and in fact when compiling a program you must tell him to include the MATLAB libraries! Also python is not very powerfull when comming to neural networks, thing that i'm now used to the simple way how matlab handles it! So if you find python powerfull go use it, but why use a script language when we can have a script language that can compile into programs well this question i will lead to you!
I'm an electrical engineering undergraduate and have written tens of thousands of lines of matlab; I would hope I know how to vectorize by now. Yes, it most certainly does use more memory because of how variables need to be set up for proper vectorization. If you you insist on proof that you can vectorize while saving both time and memory, I'll provide further evidence.
rayden wrote:
but why use a script language when we can have a script language that can compile into programs well this question i will lead to you!


...because then it isn't a scripting language Rolling Eyes Again, I think you just failed to find and use NumPy + Psycho
@Kerm you save memory agains loops but you can escape from the memory that the algorythm itself uses...

@kll you are wrong is a script because it run's on script do you know what is a M-file if not is something like a bat file, but for MATLAB. It have a feature now that can compile programs but it still is a script lang, and i have used numpy stupid morron
Rayden, when you insult people in your arguments you end up looking like a crazy person. Try to be calm when you are discussing something, because it will be much more convincing.
But it's what kll is! He is so dummy that he does not understand when we say things!
kll knows what he is talking about don't try and tell us otherwise because it has been proven many times that kll will be right and if he isn't he will admit it. but even I know enough about programing to know what you are saying is not true.
rayden wrote:
But it's what kll is! He is so dummy that he does not understand when we say things!


Maybe because you speak in some seriously broken English. I know you are foreign, but you are still responsible for communicating your point clearly.
Kllrnohj wrote:
rayden wrote:
But it's what kll is! He is so dummy that he does not understand when we say things!


Maybe because you speak in some seriously broken English. I know you are foreign, but you are still responsible for communicating your point clearly.
I was going to point out the language barrier, but I actually don't think it's playing a part in this case. It seems pretty clear what each person is saying to me: Rayden says Matlab with vectorized matlab scripting is faster and more memory-efficient than Python with NumPy, and Kllrnohj is saying they're roughly equivalent. Correct?
when compiled yes... And for the vectorized algos here it is!!!

http://www.eecs.umich.edu/dco/faq/matlab-6.5/help/techdoc/matlab_prog/ch7_per2.html

these are some but there are more
  
Register to Join the Conversation
Have your own thoughts to add to this or any other topic? Want to ask a question, offer a suggestion, share your own programs and projects, upload a file to the file archives, get help with calculator and computer programming, or simply chat with like-minded coders and tech and calculator enthusiasts via the site-wide AJAX SAX widget? Registration for a free Cemetech account only takes a minute.

» Go to Registration page
Page 3 of 3
» All times are GMT - 5 Hours
 
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

 

Advertisement