Right now the Gibbs sampler doesn't make use of many optimizations because I can't set custom compiler flags for -march=native -O3 in gcc. This is because on Windows it's going to use msvc to compile, which uses /arch:AVX for some more optimizations.
Some notes on discovering which compiler is used.
Some tests for compiler flags.
/arch documentation for msvc
Right now the Gibbs sampler doesn't make use of many optimizations because I can't set custom compiler flags for
-march=native -O3in gcc. This is because on Windows it's going to usemsvcto compile, which uses/arch:AVXfor some more optimizations.Some notes on discovering which compiler is used.
Some tests for compiler flags.
/archdocumentation for msvc