gpp_expected_improvement_gpu_test

Contents:

gpp_expected_improvement_gpu_test.hpp

Functions for testing expected improvement functions on GPU.

Tests are broken into two main groups:

  • consistency test against analytical 1,0-EI result
  • compare with CPU(MC) results

Note

These tests do not run if GPU computation (OL_GPU_ENABLED) is disabled.

namespace optimal_learning

Macro to allow restrict as a keyword for C++ compilation and CUDA/nvcc compilation. See related entry in gpp_common.hpp for more details.

gpp_expected_improvement_gpu_test.cpp

Routines to test the functions in gpp_expected_improvement_gpu.cpp.

The tests verify ExpectedImprovementGPUEvaluator from gpp_expected_improvement_gpu.cpp.

  1. Monte-Carlo EI vs analytic EI validation: the monte-carlo versions are run to “high” accuracy and checked against analytic formulae when applicable
  2. GPU EI vs CPU EI: both use monte-carlo version and run consistency check on various random sample points

namespace optimal_learning

Macro to allow restrict as a keyword for C++ compilation and CUDA/nvcc compilation. See related entry in gpp_common.hpp for more details.

Functions

int RunGPUTests()

Invoke all tests for GPU functions.

Returns:
number of test failures: 0 if all is working well.