Julien Schueller
Julien Schueller
the dimension is 0 in HSICEstimatorImplementation.cxx at line 250: `CovarianceMatrix shuffledCovariance(outputCovarianceMatrix_.getDimension());` tbb just makes it difficult to read the error, but really its just an assert (with glibcxx assertions) note...
the HSICPValuesPermutationFunctor copy ctor is missing attributes, but it shouldnt be necessary in the first place
closing stale
closing stale
I cannot reproduce, tried both gcc 13/14 && clang 18: ``` git clone --depth 1 https://github.com/stevengj/nlopt.git && cd nlopt && cmake -DCMAKE_C_COMPILER=/usr/bin/clang -DCMAKE_INSTALL_PREFIX=$PWD/install -DCMAKE_BUILD_TYPE=RelWithDebInfo -DCMAKE_C_FLAGS="-fsanitize=address,undefined" -DCMAKE_EXE_LINKER_FLAGS="-fsanitize=address,undefined" -DNLOPT_GUILE=OFF -DNLOPT_PYTHON=OFF -DNLOPT_TESTS=ON ....
saving to a file is probably not the most flexible option, instead you can instrument the fitness method in the client code
what should we see exactly ? I see vmin/vmax arguments are actually passed to matplotlib
see also #2341
began in #2944, but we should reduce the number of sections even more and improve presentation
todo: - rename functions-> functional modelling, include orthognonal basis... - match sections of examples - try toctree depth=2 - regroup doe into data analysis - regroup stochastic process inside probabilisti...