GitHub - triilman25/evaluation-machine-for-classification-models

1 min read Original article ↗

Model Evaluation Machine

Build Application from Source Code

Tools and Dependency Requirement

  • LibTorch with CPU release version (required)
  • OpenCV (required)
  • Raylib (but directly installed when cmake is configured)
  • Cmake

first of all, replace or change dependency's path of LibTorch and openCV in CMakeLists to the same path directory in your system.

    set(CMAKE_PREFIX_PATH "C:/path/to/libtorch") #Libtorch
    set(OpenCV_DIR "C:/opencv/build/x64/vc16/lib") #OpenCV

second, create build directory (more simple by using IDE like clion for the rest tutorial) on the same hierarchy with the main file or Cmake list file.

the last, get to the build directory, configure and build application by using command bellow:

cmake ..
cmake --Build . -DCMAKE_BUILD_TYPE=Release

Shortcut

This application, especially version 1.0.5 is not properly built with rich features. The application doesn't include manual instructions in it. You can use the shortcut bellow:

  • ctrl_left + [left click] - erasing all input character in textbox.
  • shift + [1/2] - switching between light and dark mode.
  • scroll - using scroll for increase and decrease input size instead just using arrows.

NOTE : You can drag and drop file in the box area of textbox directly.

Showcase

Light Mode Dark Mode
image1 image2
image3 image4

Credit:

The icon used in this application can be found on www.kenney.nl