Skip to content

Architecturally optimized neural networks trained with regularized backpropagation

Notifications You must be signed in to change notification settings

dicarlolab/archconvnets

Repository files navigation

archconvnets

Architecturally optimized convolutional neural networks trained with regularized backpropagation

install

git clone this repository and add the path to the PYTHON_PATH variable

follow the install instructions for all requirements listed in requirements.txt (including the requirements in those requirements files)

you have to download the cifar-10 dataset, and then set the environment variable CIFAR10_PATH to its location (untarred) to run tests properly:

cd ~/.skdata
wget http://www.cs.toronto.edu/~kriz/cifar-10-py-colmajor.tar.gz
export CIFAR10_PATH=~/.skdata/cifar-10-py-colmajor

Install CUDA: http://sn0v.wordpress.com/2012/12/07/installing-cuda-5-on-ubuntu-12-04/

If you're on a machine other than honeybadger (or one that is similarly configured) modify archconvnets/convnet/build.sh to match your machine's setup (cuda, python, and numpy locations must be specified)

Compile

sh build.sh

running tests

you must be in the archconvnets/convnet directory to run tests:

nosetests tests

change build to match your local settings (cuda location, python location, numpy location)

sh build.sh

There is more extensive documentation at https://code.google.com/p/cuda-convnet/ which forms the convolutional neural network "backend"

About

Architecturally optimized neural networks trained with regularized backpropagation

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •