Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

BenchMarking for Codegen and CodegenNoGrad() #282

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

BenchMarking for Codegen and CodegenNoGrad() #282

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions root/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,10 +1,10 @@
add_subdirectory(interpreter)
add_subdirectory(io)
#add_subdirectory(io)
add_subdirectory(hist)
add_subdirectory(math)
add_subdirectory(pyroot)
if (roofit)
add_subdirectory(roofit)
endif()
add_subdirectory(tree)
#add_subdirectory(tree)
add_subdirectory(tmva)
12 changes: 12 additions & 0 deletions root/roofit/roofit/RooFitUnBinnedBenchmarks.cxx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -239,13 +239,20 @@ const auto unit = benchmark::kMillisecond;
auto Legacy = static_cast<int>(RooFit::EvalBackend::Value::Legacy);
auto Cpu = static_cast<int>(RooFit::EvalBackend::Value::Cpu);
auto Cuda = static_cast<int>(RooFit::EvalBackend::Value::Cuda);
//testing for codegen and codegen_no_grad-codegen without clad:
auto Codegen = static_cast<int>(RooFit::EvalBackend::Value::Codegen);
auto CodegenNoGrad = static_cast<int>(RooFit::EvalBackend::Value::CodegenNoGrad);


#define ARGS UseRealTime()->Unit(unit)

BENCHMARK(BDecayWithMixing)->Name("BDecayWithMixing_FitLegacy")->Args({nEvents, Legacy, 1})->ARGS;
BENCHMARK(BDecayWithMixing)->Name("BDecayWithMixing_FitLegacyNumCPU2")->Args({nEvents, Legacy, 2})->ARGS;
BENCHMARK(BDecayWithMixing)->Name("BDecayWithMixing_FitLegacyNumCPU4")->Args({nEvents, Legacy, 4})->ARGS;
BENCHMARK(BDecayWithMixing)->Name("BDecayWithMixing_FitCPU")->Args({nEvents, Cpu, 1})->ARGS;
//testing for codegen and codegennograd
BENCHMARK(BDecayWithMixing)->Name("BDecayWithMixing_FitCodegen")->Args({nEvents, Codegen, 1})->ARGS;
BENCHMARK(BDecayWithMixing)->Name("BDecayWithMixing_FitCodegenNoGrad")->Args({nEvents, CodegenNoGrad, 1})->ARGS;
#ifdef DO_BENCH_ROOFIT_CUDA
BENCHMARK(BDecayWithMixing)->Name("BDecayWithMixing_FitCUDA")->Args({nEvents, Cuda, 1})->ARGS;
#endif
Expand All @@ -254,6 +261,9 @@ BENCHMARK(BDecayGaussResolution)->Name("BDecayGaussResolution_FitLegacy")->Args(
BENCHMARK(BDecayGaussResolution)->Name("BDecayGaussResolution_FitLegacyNumCPU2")->Args({nEvents, Legacy, 2})->ARGS;
BENCHMARK(BDecayGaussResolution)->Name("BDecayGaussResolution_FitLegacyNumCPU4")->Args({nEvents, Legacy, 4})->ARGS;
BENCHMARK(BDecayGaussResolution)->Name("BDecayGaussResolution_FitCPU")->Args({nEvents, Cpu, 1})->ARGS;
//testing for codegen and codegennograd
BENCHMARK(BDecayGaussResolution)->Name("BDecayGaussResolution_FitCodegen")->Args({nEvents, Codegen, 1})->ARGS;
BENCHMARK(BDecayGaussResolution)->Name("BDecayGaussResolution_FitCodegenNoGrad")->Args({nEvents, CodegenNoGrad, 1})->ARGS;
#ifdef DO_BENCH_ROOFIT_CUDA
BENCHMARK(BDecayGaussResolution)->Name("BDecayGaussResolution_FitCUDA")->Args({nEvents, Cuda, 1})->ARGS;
#endif
Expand All @@ -262,6 +272,8 @@ BENCHMARK(BDecayDoubleGauss)->Name("BDecayDoubleGauss_FitLegacy")->Args({nEvents
BENCHMARK(BDecayDoubleGauss)->Name("BDecayDoubleGauss_FitLegacyNumCPU2")->Args({nEvents, Legacy, 2})->ARGS;
BENCHMARK(BDecayDoubleGauss)->Name("BDecayDoubleGauss_FitLegacyNumCPU4")->Args({nEvents, Legacy, 4})->ARGS;
BENCHMARK(BDecayDoubleGauss)->Name("BDecayDoubleGauss_FitCPU")->Args({nEvents, Cpu, 1})->ARGS;
BENCHMARK(BDecayDoubleGauss)->Name("BDecayDoubleGauss_FitCodegen")->Args({nEvents, Codegen, 1})->ARGS;
BENCHMARK(BDecayDoubleGauss)->Name("BDecayDoubleGauss_FitCodegenNoGrad")->Args({nEvents, CodegenNoGrad, 1})->ARGS;
#ifdef DO_BENCH_ROOFIT_CUDA
BENCHMARK(BDecayDoubleGauss)->Name("BDecayDoubleGauss_FitCUDA")->Args({nEvents, Cuda, 1})->ARGS;
#endif
Expand Down
20 changes: 17 additions & 3 deletions root/roofit/roofit/benchRooFitBackends.cxx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,11 +22,11 @@
#include <benchmark/benchmark.h>

int printLevel = 0;
size_t nEvents = 100000;
size_t nEvents = 10000;
const auto minimizerName = "Minuit2";
constexpr bool verbose = false;

enum RunConfig_t { runScalar, runCpu, fitScalar, fitCpu, fitCuda };
enum RunConfig_t { runScalar, runCpu, fitScalar, fitCpu, fitCuda};//, fitCodegen, fitCodegenNoGrad };

void runFitBenchmark(benchmark::State &state, RooAbsPdf &pdf, RooAbsData &data)
{
Expand All @@ -45,7 +45,11 @@ void runFitBenchmark(benchmark::State &state, RooAbsPdf &pdf, RooAbsData &data)
pdf.fitTo(data, EvalBackend::Cpu(), Minimizer(minimizerName), PrintLevel(printLevel - 1), PrintEvalErrors(-1));
} else if (runConfig == fitCuda) {
pdf.fitTo(data, EvalBackend::Cuda(), Minimizer(minimizerName), PrintLevel(printLevel - 1), PrintEvalErrors(-1));
}
}// else if (runConfig == fitCodegen) {
// pdf.fitTo(data, EvalBackend::Codegen(), Minimizer(minimizerName), PrintLevel(printLevel - 1), PrintEvalErrors(-1));
// } else if (runConfig == fitCodegenNoGrad) {
// pdf.fitTo(data, EvalBackend::CodegenNoGrad(), Minimizer(minimizerName), PrintLevel(printLevel - 1), PrintEvalErrors(-1));
// }
state.PauseTiming();
params.assign(paramsInitial);
state.ResumeTiming();
Expand Down Expand Up @@ -332,25 +336,35 @@ auto const unit = benchmark::kMillisecond;
BENCHMARK(benchFitGauss)->Unit(unit)->Name("Gaus_FitLegacy")->Args({fitScalar});
BENCHMARK(benchFitGauss)->Unit(unit)->Name("Gaus_FitCPU")->Args({fitCpu});
CUDA_ONLY(BENCHMARK(benchFitGauss)->Unit(unit)->Name("Gaus_FitCUDA")->Args({fitCuda}));
BENCHMARK(benchFitGauss)->Unit(unit)->Name("Gaus_FitCodegen")->Args({fitCodegen});
BENCHMARK(benchFitGauss)->Unit(unit)->Name("Gaus_FitCodegen_No_Grad")->Args({fitCodegenNoGrad});

BENCHMARK(benchFitGaussXSigma)->Unit(unit)->Name("GausXS_FitLegacy")->Args({fitScalar});
BENCHMARK(benchFitGaussXSigma)->Unit(unit)->Name("GausXS_FitCPU")->Args({fitCpu});
CUDA_ONLY(BENCHMARK(benchFitGaussXSigma)->Unit(unit)->Name("GausXS_FitCUDA")->Args({fitCuda}));
BENCHMARK(benchFitGaussXSigma)->Unit(unit)->Name("GausXS_FitCodegen")->Args({fitCodegen});
BENCHMARK(benchFitGaussXSigma)->Unit(unit)->Name("GausXS_FitCodegen_No_Grad")->Args({fitCodegenNoGrad});

BENCHMARK(benchFit)->Unit(unit)->Name("AddPdf_FitLegacy")->Args({fitScalar});
BENCHMARK(benchFit)->Unit(unit)->Name("AddPdf_FitCPU")->Args({fitCpu});
CUDA_ONLY(BENCHMARK(benchFit)->Unit(unit)->Name("AddPdf_FitCUDA")->Args({fitCuda}));
BENCHMARK(benchFit)->Unit(unit)->Name("AddPdf_FitCodegen")->Args({fitCodegen});
BENCHMARK(benchFit)->Unit(unit)->Name("AddPdf_FitCodegen_No_Grad")->Args({fitCodegenNoGrad});

BENCHMARK(benchProdPdf)->Unit(unit)->Name("ProdPdf_FitLegacy")->Unit(benchmark::kMillisecond)->Args({fitScalar});
BENCHMARK(benchProdPdf)->Unit(unit)->Name("ProdPdf_FitCPU")->Unit(benchmark::kMillisecond)->Args({fitCpu});
CUDA_ONLY(BENCHMARK(benchProdPdf)->Unit(unit)->Name("ProdPdf_FitCUDA")->Unit(benchmark::kMillisecond)->Args({fitCuda}));
BENCHMARK(benchProdPdf)->Unit(unit)->Name("ProdPdf_FitCodegen")->Unit(benchmark::kMillisecond)->Args({fitCodegen});
BENCHMARK(benchProdPdf)->Unit(unit)->Name("ProdPdf_FitCodegen_No_Grad")->Unit(benchmark::kMillisecond)->Args({fitCodegenNoGrad});

// Watch out with the result from these benchmarks: if there are evaluation
// errors during the minimization, the time differences will mostly come from
// how the backends are handling those. This might or might not be wanted.
BENCHMARK(benchModel)->Unit(unit)->Name("FitModel_FitLegacy")->Unit(benchmark::kMillisecond)->Args({fitScalar});
BENCHMARK(benchModel)->Unit(unit)->Name("FitModel_FitCPU")->Unit(benchmark::kMillisecond)->Args({fitCpu});
CUDA_ONLY(BENCHMARK(benchModel)->Unit(unit)->Name("FitModel_FitCUDA")->Unit(benchmark::kMillisecond)->Args({fitCuda}));
BENCHMARK(benchModel)->Unit(unit)->Name("FitModel_FitCodegen")->Unit(benchmark::kMillisecond)->Args({fitCodegen});
BENCHMARK(benchModel)->Unit(unit)->Name("FitModel_FitCodegen_No_Grad")->Unit(benchmark::kMillisecond)->Args({fitCodegenNoGrad});

int main(int argc, char **argv)
{
Expand Down
Loading