some optimizations and corrections to ctr kernels, that don't yet fix the bug...
Edgar Solomonik authored
some optimizations and corrections to ctr kernels, that don't yet fix the bug raised by libtensor tests
d4568709
Name Last commit Last update
bench wrote draft of global ordering function, combined reuse_buffers if statement to be more compact
bin Added .gitignore files to bin and lib
doc Conflicts should have already been previously resolved in memeff_redist, so no significant changes made in this merge
examples Added proper diagonal rescaling and integrated it into contractions so they no longer need to be unpacked for preserved and summed SY groups.
include Conflicts should have already been previously resolved in memeff_redist, so no significant changes made in this merge
lib Added .gitignore files to bin and lib
src some optimizations and corrections to ctr kernels, that don't yet fix the bug raised by libtensor tests
studies Conflicts should have already been previously resolved in memeff_redist, so no significant changes made in this merge
test Added proper diagonal rescaling and integrated it into contractions so they no longer need to be unpacked for preserved and summed SY groups.
.gitignore Added .gitignore files to bin and lib
Makefile 1. Added capability (via constructor) to repack tensor data into a different packed symmetric layout without doing the normal symmetrization permutations that happen during sum
README.md back to normal README
configure include the INCLUDES
license.txt Got rid of the unbearably ugly copy right warning from all the files and put in a shortened one.

A distributed memory library for efficient decomposition of tensors of arbitrary type and parallel MPI+OpenMP execution of tensor contractions/functions.

Full documentation is available at

http://ctf.cs.berkeley.edu

Basic build process: ./configure; make;

Basic test process: sequential: 'make test', parallel: 'make test2'