CTorch is a C++-based tool designed for training and testing neural networks. The project leverages the Eigen library for efficient matrix operations and mathematical computations. It aims to provide a high-performance environment for neural network development.
Inspired by the Deep Learning project.
-
C++ Standard: Requires support for C++20 or later.
-
Eigen Library: The Eigen library must be downloaded and placed in the root directory of the project workspace.
- Download Eigen: You can download Eigen 3.4.0 release at Eigen_3.4.0
- Extract Eigen: Extract the downloaded
.zip
or.tar.gz
file. - Place Eigen: Copy the
Eigen
in extracted foldereigen-3.4.0
to Group-86 workspace(Ensure that you are in sprint1 branch). The structure should look like this:Group-86/ ├── CMakeLists.txt ├── Data ├── datasets ├── DataType ├── Eigen/ # Place the extracted Eigen library here └── ...
- GCC: Install the latest version of GCC using your package manager. For example:
Verify the installation:
sudo apt update sudo apt install g++-10
g++ --version
- Clang: Alternatively, you can install Clang:
sudo apt install clang
- Xcode: Install Xcode from the App Store or use the command line tools:
Ensure your compiler supports C++20.
xcode-select --install
- Visual Studio: Install the latest version of Visual Studio and select the C++ development workload. Ensure that the MSVC compiler supports C++20.
- MSYS2: Alternatively, install GCC through MSYS2:
pacman -S mingw-w64-x86_64-gcc
Run the following commands in a terminal to build and run the project:
mkdir build
cd build
cmake ..
make
After successful compilation, execute the following command to run the program:
./CTorch
You can see the example output of the dummy network at the terminal:
For any issues, please submit an issue or contact us via email.