Ryan Dickie wrote:

I also hate matlab to death. Is there any possibility of using haskell as a replacement using ghci?

Yes. The strongly typed linear algebra project (Vectro) does exactly that. With an added guarantee that attempting to add or multiply matrices of inappropriate sizes is a type error. Thus typical errors like forgetting to transpose a matrix/vector are reported before the computation starts. The paper shows a few examples, including a larger example of a learning algorithm. http://ofb.net/~frederik/stla/