Home

Awesome

Prototype for lung nodule detection

This repository contains multiple useful scripts to train (and score) models with mxnet for the LUNA challenge.

Dependencies

First, add this library to your python path. Then install the following libraries if they are not already installed:

Libraries

Install these with pip:

pip install -r requirements.txt

mxnet

Either install manually (needed to enable GPU support) or use pip:

pip install mxnet

OpenCV

Either install manually or use pip:

pip install opencv-python

Add to python path

Add the scripts path to your PYTHONPATH.

If you use virtualenv you can use a .pth file, e.g. execute this command from the git root repository (replace [YOUR_VENV_PATH]):

echo "$(pwd)/scripts" > [YOUR_VENV_PATH]/lib/python2.7/site-packages/lucad.pth

Otherwise add something like this to your .bashrc or .zshrc and logout/login or reboot:

export PYTHONPATH="$PYTHONPATH:[GIT_ROOT]/scripts"

Execute scripts

Please execute all scripts from this directory, i.e. to start the viewer use this command from git root directory:

python scripts/viewer/viewer.py path/to/LUNA/data