Implement feedforward, convolution neural network in C++ with only built-in libraries. This is a tutorial style project which implements back propagation in neural networks without using external libraries. A 2D Tensor is defined and all the required Linear Algebra operations are defined to perform required computations. Both the networks are tested on mnist dataset which is attached along the project.
psrikanthm/cnn-from-scratch
Implement feedforward, convolution neural network in C++ with only built-in libraries. This is a tutorial style project which implements back propagation in neural networks without using external libraries. A 2D Tensor is defined and all the required Linear Algebra operations are defined to perform required computations. Both the networks are tested on mnist dataset which is attached along the project.
C++MIT