intel-mkl

There was a post on this here: https://gist.github.com/JonathanRaiman/f2ce5331750da7b2d4e9 which shows a great speed improvement by just calling the Fortran libraries (BLAS / LAPACK / Intel MKL / OpenBLAS / whatever you installed with NumPy). After many hours of working on this (because of deprecated SciPy libraries) I finally got it to compile with no results. It was 2x faster than NumPy. Unfortunately as another user pointed out, the Fortran routine is always adding the output matrix to the new results calculated, so it only matches NumPy on the 1st run. I.e. A := alpha*x*y.T + A . So that

I'm using NumPy built against Intel's Math Kernel Library. I use virtualenv, and typically use pip to install packages. However, in order for NumPy to find the MKL libraries, it's necessary to create a site.cfg file in the NumPy source directory prior to compiling it, then manually build and install. I could script this whole process, but I was hoping for a simpler solution. I have a standard site.cfg file that can be used for this purpose under version control. Are there any pip command line options that will tell it to copy a particular file to the source directory before building a package?

I want to call mkl.mkl_scsrmultcsr from python. The goal is to calculate a sparse matrix C in compressed sparse row format. Sparse matrix C is the matrix product between A and transpose of A, where A is also a sparse matrix in csr format. When calculating C = A dot (A.T) with scipy, scipy seems (?) to allocate new memory for holding transpose of A (A.T), and definitely allocates memory for a new C matrix (This means I can't use an existing C matrix). So, I want to try to use the mkl c function directly to decrease memory usage. Here is an answer that works for another mkl function. In that