blas

This has to deal with a similar problem here: Calling BLAS / LAPACK directly using the SciPy interface and Cython but is different because I'm using the actual code in the SciPy example here _test_dgemm : https://github.com/scipy/scipy/blob/master/scipy/linalg/cython_blas.pyx which is extremely fast (5x faster than numpy.dot when an output matrix is input or 20x faster if not). It produces no results if Mx1 1xN vectors are passed. It produces the same values as numpy.dot with matrices passed. I've minimized the code since no answers have been posted for clarity. Here's the dgemm.pyx. : import

When I set a ulimit -v on a powerful machine I'm working on (48 × Intel(R) Xeon(R) CPU E7-4860 v2 @ 2.60GHz, 2 TB RAM, Red Hat Enterprise Linux Server release 6.7), python fails to import numpy when I'm working in my virtualenv, as shown below. Everything seems fine when I impose no resource limitations or when I'm not working inside virtualenv. I suppose this is a bug somewhere. Did I set up something wrongly? Where is the correct place to report this? NumPy? OpenBLAS? System administrator? $ ulimit -a core file size (blocks, -c) 0 data seg size (kbytes, -d) unlimited scheduling priority (-e)

The BLAS section in R installation and administration manual says that when R is built from source , with configuration parameter --without-blas , it will build Netlib's reference BLAS into a standalone shared library at R_HOME/lib/libRblas.so , along side the standard R shared library R_HOME/lib/libR.so . This makes it easier for user to switch and benchmark different tuned BLAS in R environment. The guide suggests that researcher might use symbolic link to libRblas.so to achieve this, and this article gives more details on this. On contrary, when simply installing a pre-compiled binary

An algorithm I'm working on requires computing, in a couple places, a type of matrix triple product. The operation takes three square matrices with identical dimensions, and produces a 3-index tensor. Labeling the operands A , B and C , the (i,j,k) -th element of the result is X[i,j,k] = \sum_a A[i,a] B[a,j] C[k,a] In numpy, you can compute this with einsum('ia,aj,ka->ijk', A, B, C) . Questions: Does this operation have a standard name? Can I compute this with a single BLAS call? Are there any other heavy-optimized numerical C/Fortran libraries that can compute expressions of this type?