On Tue, Sep 10, 2019 at 04:18:16AM +0000, Hearn, David J. wrote: > Hi Ryan, > > Thank you very much for your help! > > In case this is of use, I ended up removing my new boost installation, > reinstalling an older boost (using yum), switching to an older gcc > (4.8.5), and using the following cmake: > > cmake ../ -DCMAKE_INSTALL_PREFIX:PATH=/usr/local/software/mlpack > -DUSE_OPENMP=OFF > -DARMADILLO_INCLUDE_DIR=/usr/local/software/armadillo-9.600.6/include/ > > -DARMADILLO_LIBRARY=/usr/local/software/armadillo-9.600.6/lib64/libarmadillo.so > -DBUILD_PYTHON_BINDINGS=OFF -DBoost_NO_BOOST_CMAKE=1 > -DBUILD_TESTS=OFF
This seems like a reasonable configuration step to get things working right, given that Armadillo is in /usr/local/software/ and not in the default search path. > No matter what I tried, gcc 9.2 would not finish the compilation. Hmm, can you provide the error message? Also, was this the git master branch or the latest release (3.1.2)? > The reuters_example.cpp would not compile unless I commented out: > > CLI::ParseCommandLine(argc, argv); Where did you find these examples? If the docs are out of date I can take a minute and update them, but I don't know about the Reuters example. > Now the mlpack-examples seem to compile and run without a core dump, > so this is definite progress, although the output from knn_example > appears to be different from the provided "expected" output. I'm a bit confused too on this one---where is the 'expected' output for the KNN example? When I read this page: http://mlpack.org/doc/mlpack-3.1.1/doxygen/sample.html it doesn't specify particular expected output. > Many thanks again! I think I'm to a place where I can begin using > mlpack. Great! :) Thanks, Ryan -- Ryan Curtin | "He's a peculiar man. You could even say that he [email protected] | has principles." - Carson Wells _______________________________________________ mlpack mailing list [email protected] http://knife.lugatgt.org/cgi-bin/mailman/listinfo/mlpack
