Home

Awesome

Deep Residual Learning for Image Recognition

This is a pytorch implementation of "Deep Residual Learning for Image Recognition",Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun the winners of the 2015 ILSVRC and COCO challenges.

It's forked from Michael Wilber's torch-residual-networks . The data loading and preprocessing have been moved from the lua side into the python side, so you can easily modify the data loading and preprocessing, using the python tools and libraries you're used to using.

For full readme on the original torch-residual-networks library, please see https://github.com/gcr/torch-residual-networks/network

How to use

git clone https://github.com/torch/distro.git ~/torch --recursive
cd ~/torch
# install dependencies.  To install everything:
  bash install-deps
# Or, if you're on ubuntu, you only need the following dependencies:
  sudo apt-get update -y
  sudo apt-get install -y wget git gcc g++ cmake libffi-dev \
       libblas-dev liblapack-dev libatlas-base-dev gfortran libreadline-dev
# install torch
./install.sh
luarocks install cudnn
luarocks install nninit
sudo apt-get install python2.7-dev
virtualenv -p python2.7 ~/env27
source ~/env27/bin/activate
pip install docopt
pip install numpy
git clone https://github.com/hughperkins/pytorch ~/pytorch
cd ~/pytorch
source ~/torch/install/bin/torch-activate
./build.sh
git clone https://github.com/hughperkins/pytorch-residual-networks ~/pytorch-residual-networks
cd ~/pytorch-residual-networks

Possible issues, and how to deal with them

Changes

2016 April 12:

2016 April 11: