class Lightgbm < Formula desc "Fast, distributed, high performance gradient boosting framework" homepage "https://github.com/microsoft/LightGBM" url "https://github.com/microsoft/LightGBM/archive/v3.0.0.tar.gz" sha256 "940563ece81d94eb843c2d1dc702da5af7c2a67fb1ccf31d64b39e00262d35b5" license "MIT" bottle do cellar :any sha256 "24a02453d33224cbe3f0d5f2f623d1356eb4b26c6bb6367e0da4c644ffaedd60" => :catalina sha256 "681eaac747cc86830db2d7162715803a794b95678af3d9072dafa80483ee6699" => :mojave sha256 "8cbaa97be86257f42bcaecfc1b8ee805558351e11438976606f9f58a897ed23d" => :high_sierra end depends_on "cmake" => :build depends_on "libomp" def install mkdir "build" do system "cmake", *std_cmake_args, "-DAPPLE_OUTPUT_DYLIB=ON", ".." system "make" system "make", "install" end pkgshare.install "examples" end test do cp_r (pkgshare/"examples/regression"), testpath cd "regression" do system "#{bin}/lightgbm", "config=train.conf" end end end