A Deep Learning Framework in C++, developed with Just-in-time Compilation and Symbolic Computation.
IntellGraph is an abbreviation of Intelligent Graph. As the name indicates, the IntellGraph framework is developed for Artifical Intelligence and is abstracted based on Graph Theory. The project is still under development. In current version, users are able to use it for constructing fully connected deep neural networks with different activation and loss functions (e.g. sigmoid activation function, mean square error loss function, cross-entropy loss function, etc). Examples (in the example/ directory) are prepared to show the capability of the IntellGraph project and you are encouraged to study them before building your own neural networks.
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.
First we need to check out the git repo:
$ cd ${insert your workspace folder here}
$ git clone https://github.com/lingbozhang/IntellGraph my-project
Now we should be in the project's top level folder.
$ rm -rf build/manual && mkdir -p build/manual
$ cd build/manual
$ conan install ../..
$ cmake ../..
$ make && make install
To run examples (codes are located in the examples/ directory), do following:
$ cd intellgraph/build/manual/bin
$ ./examples
We welcome every people who interested in the IntellGraph project, if you want to contribute, please do not hesitate to contact us [email protected]. You are recommended to review the contribution guidelines and development tutorial. By participating, you are also expected to join the Gitter community:
- IntellGraph uses and modifies the cmake-project-template developed by Konstantin Gredeskoul.