OpenANN  1.1.0
An open source library for artificial neural networks.
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Pages
Installation

Requirements

At the moment we support only the operating systems Linux and MacOS and the compilers GCC and Clang.

The following programs and libraries are required to use OpenANN:

Compilation Documentation Examples Python bindings Benchmarks
  • OpenGL
  • GLUT
  • libxmu
  • Qt4
  • Python (header files)
  • NumPy
  • Cython
  • Python
  • Matplotlib
  • SciPy

Installation

Go to the OpenANN directory and execute

mkdir build
cd build
cmake -D CMAKE_BUILD_TYPE=Release ..
sudo make install
sudo ldconfig

Build Configuration

The OpenANN build is influenced by some options. These can be set with cmake:

cmake -D <option>=<value> ..

You will see an example in the previous section.

Option Description Default
CMAKE_BUILD_TYPE Build configuration. Available options are: Debug, Release. Release
LOG_LEVEL Log level of the internal logging. Available options are: DISABLED, ERROR, INFO, DEBUG INFO
PARALLEL_CORES Number of CPU cores that OpenANN can use. 4
ALWAYS_BUILD_DOCUMENTATION Add documentation target to default target (i.e. "make all"). OFF
BUILD_PYTHON (De)activate building Python bindings for OpenANN. ON
EXCLUDE_TESTS_FROM_ALL You can build all examples, benchmarks and the test suite with the standard build target when you deactivate this. This might be useful for developers. ON