how to speed up the process

 my dataset is aroud 32000rows and 6 columns

my program link

http://www.mathworks.com/matlabcentral/fileexchange/24343-discretization-algorithms-class-attribute-contingency-coefficient

it takes 24 hrs to run,please help how to speed up the process

Answer 1

[Jan]

At first search for the most time consuming part of the function using the PROFILEr. It would be useless to optimize a part, which takes 0.1% of the processing time.

Then examine the found "bottleneck" and check for a proper pre-allocation, the possibility to use integer types instead of doubles, simplify the contents of loops and avoid all repeated calculations. Finally check if a vectorization could be possible.

Does the algorithm exhaust your RAM? Then installing enough memory can give a speed-up of factor 1000.

Can the problem be parallelized? Then PARFOR might be helpful.

After or before the above steps have been performed, buy a computer with the double speed. Then the computation will take 50% of the time.

[Pat]

wen my program runs ,it is in busy state ,dont know whether it is running or hanged off,is it possible to show which line it is executing

[Walter Roberson]

In order to show the line that is executing, you would need to use a timer interrupt that examined dbstack() and printed out that information.

[Jan]

The "echo" command will show the currently processed line. See "help echo". Take into account, that this will slow down the processing dramatically.

Instead of this brute display of every line, you could use "waitbar" or some "disp" statements to show the progess.

[Pat]

OriginalData=xlsread('dataset.xls');

[r c]=size(OriginalData)

k=zeros(r,1);

OriginalData=[OriginalData k];

[ DiscretData,DiscretizationSet1 ] = CACC_Discretization( OriginalData,1);

DiscretizationSet1

this is my program can u please tell where to use those commands to see which line is ececuting