Skip to content

Latest commit

 

History

History
41 lines (40 loc) · 1.02 KB

README.md

File metadata and controls

41 lines (40 loc) · 1.02 KB

MazeNet

Mnist multi perceptron neural-network training from scratch with c++ and opencv matrix.
in this repository there are two folder that includes folder contains header files and src contains cpp programs.

How to train

First clone repository via this command :

git clone https://github.com/mahdizynali/MazeNet.git

Now try to run shell script easy to use :

chmod a+x run.sh
./run.sh

Run as developer

after that you have to extract dataset from compress file (dataset.tar.xz). tar.xz is being support by linux & mac systems.
you would change config.hpp header in order to set your hyperparameter as you need.

# define l_rate 0.005
# define total_epochs 10
# define batch_size 64

right now inside MazeNet folder follow these commands to build and run training loop :

mkdir build && cd build
make

Or try this for a bit faster :

make -j`nproc`

finally run the program :

./maze

after training , neuron weights will recive and save into a yml file besides maze program.