Also, I am a novice in open source development, so, I would really appreciate your help and direction.
On Wed, Mar 28, 2012 at 8:29 PM, Monalisa Gupta <monalisa....@gmail.com>wrote: > My apologies for all the delay and no reply from my side. Was caught up > with exams. > > I read the https://github.com/sympy/sympy/wiki/Linear-Algebra-Vision and > https://github.com/sympy/sympy/wiki/Linear-Algebra-Module. Can you please > give me a bit more idea of what functions need to be implemented for tensor > algebra and linear algebra? Basically, I want to know what is expected from > me in this project (if I do one :) )? What functions and applications of > tensor algebra and linear algebra are you targeting for this summer? > > On Fri, Mar 16, 2012 at 2:58 PM, krastanov.ste...@gmail.com < > krastanov.ste...@gmail.com> wrote: > >> > One possible separation of work within this category would be the >> following >> > >> > 1) Tensor core - how do we represent and simplify indexed expressions? >> How >> > do we interact with the rest of SymPy? >> > 2) Abstract Tensors - There is some serious math behind these things. >> Lets >> > write it down - I think this is where Stefan is particularly interested >> Yes, this is the project I am interested in, but I will probably start >> with abstract linear algebra before (if at all) starting work on the >> multilinear case. >> > 3) Numeric ND-Arrays - we often have a basis and want to write down >> these >> > things explicitly. There a number of concerns here (mutability, >> sparsity, >> > ground-types, numpy, code generation). >> >> > Question for Stefan: Do you think it is feasible to separate out (1) >> and (2) >> > into separate GSoC projects? >> It seems possible, but if two such projects are admitted, there should >> definitely be a clear plan for how to couple them. >> >> > There is obviously enough work involved for two >> > projects but I wonder how connected their development will have to be. >> Clear plan for the architecture and doctests should be enough. >> >> > Also, >> > where is your current tensor work? I'd love to see how you're >> formulating >> > things. I started playing with this a TensorExpr idea a while ago. My >> work >> > is here >> > https://github.com/mrocklin/sympy/tree/tensor/sympy/tensor >> Mine is just a quick and ugly hack over symbol, that I needed to >> create in a few hours. It is at the same time SymbolicTensor and >> ImmutableTensor, to borrow the naming conventions you used for the >> matrices. I will put it on github over the weekend. >> >> > As an aside, a question of mine on scicomp.stackexchange.com related >> to this >> > question was just given a bounty today. This question has the highest >> ratio >> > between question votes and answer votes that I've ever seen. A sure sign >> > that we should work on it and provide the answer. >> > >> http://scicomp.stackexchange.com/questions/74/symbolic-software-packages-for-matrix-expressions >> I think that I should be able to answer this question if my project >> succeeds. But it would be about linear operators, not matrices (it >> would be independent of the matrix representation). >> >> -- >> You received this message because you are subscribed to the Google Groups >> "sympy" group. >> To post to this group, send email to sympy@googlegroups.com. >> To unsubscribe from this group, send email to >> sympy+unsubscr...@googlegroups.com. >> For more options, visit this group at >> http://groups.google.com/group/sympy?hl=en. >> >> > > > -- > Monalisa Gupta > > -- Monalisa Gupta -- You received this message because you are subscribed to the Google Groups "sympy" group. To post to this group, send email to sympy@googlegroups.com. To unsubscribe from this group, send email to sympy+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/sympy?hl=en.
