Third as a result of my second piece of evidence its functions are somewhat lagging. Second, it isn't frequently updated, the latest version in the software centre, was last updated in 2008. My evidence for this is three part: firstly, in the software centre it receive the lowest star rating of the four software in question. It has some differences in terms of syntax but nothing you can't get your head around if you search the documentation that comes with it.įreeMat is the worst, in my opinion at least, of software that's syntax is heavily based on MATLAB. Scilab is an adequate replacement for MATLAB too. QtOctave is a frontend (GUI) for the GNU Octave programming language. I don't know if you're aware of QtOctave. The only differences, I'm aware of at least, aside from the odd toolbox, which you can attempt to replace by the packages here is that GNU Octave can define functions from the command line. Its syntax has an extreme compatibility with that of MATLAB. FreeMat and Spyder I do not have too much experience with, but I have some. If you don’t know how to use it just type “edit cov” in Matlab and then copy the whole function as a new function to your freemath program I don’t know if it’s acceptable or not but of course it works.Well you're in for a treat I have used all the programming languages/software you mentioned. clusterGeneration package:, for generating multidimensional covariance matrix.įor calculating covariance in Freemath as I don’t want to recreate the wheel I used cov function in Matlab. elliptic package:, for integral with complex numbers.ĥ. mvtnorm package:, for multivariate normal distribution.Ĥ. rnorm,dnorm,pnorm:, for dealing with 1D normal distribution.ģ. sample:, for randomly generating data from different classes.Ģ. Other than that, I'll list the packages/functions here that I think would be useful for finishing this homework in R.ġ. using prod function together with integrate). Until now, I'm still having problems with drawing the decision boundary in R and calculating the Characteristic function of F_T(\omega) in a simple way (i.e. R has more useful packages compared with FreeMat, but not as powerful as Matlab for general purpose scientific computing, except in the aspect of Statistic computing. Compared with FreeMat and Matlab, I think R is in between of them. R packages useful for this homework assignmentĭon't know yet anybody is using R for this homework. In my Rhea dropbox, as the peer review system is not quite ready yet. where do we turn this thing in? Do we upload it to Mimi's dropbox? Anybody know? Thanks! The transformed data, using either A or B, should have be the desired statistics (please verify!). Then $ B = u \sqrt v $ would serve as the square root. Perform singular value decomposition using FreeMat's "svd" command on the covariance matrix to get. Instead of starting with the covariance matrix and taking the square root, start with the upper triangular matrix A and take A'A as the covariance.To make matters worse, FreeMat cannot perform Cholesky decomposition. The solution is to generate independent standard normal data points and perform a linear transformation. I don't know if anyone else ran into this issue, but FreeMat doesn't know how to generate MVN random samples. Generating correlated multi-variate normal (MVN) data: Here is a link for a theoretical and practical assignment on Bayes Classifier. Here is a link to a lab on Bayes Classifier that you might find helpful. The next homework will consists in a peer review of this homework. This homework (see original statement here) is a freeform exercise that consists in applying Bayes decision rule to Normally distributed data. ECE662 Spring 2010, hw1 related discussions
0 Comments
Leave a Reply. |