No description, website, or topics provided.
Switch branches/tags
Nothing to show
Clone or download
Type Name Latest commit message Commit time
Failed to load latest commit information.
code remove wht dependency Dec 28, 2016
data data Jun 9, 2016
predictions hawkes pred Feb 11, 2016
.gitignore merge Jun 9, 2016 Create Dec 27, 2016



Tested under Ubuntu 14.04

Download and install cuda from
sudo dpkg -i cuda-repo-ubuntu1404_8.0.44-1_amd64.deb
sudo apt-get update
sudo apt-get install cuda

in .bashrc, add the following path (suppose you installed to the default path)

export CUDA_HOME=/usr/local/cuda
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
export LD_LIBRARY_PATH=/usr/local/lib:$LD_LIBRARY_PATH
Download and install intel mkl

in .bashrc, add the following path

source {path_to_your_intel_root/name_of_parallel_tool_box}/bin/
export MKL_ROOT={path_to_your_intel_root}/mkl
Install cppformat (now called fmtlib)
check for help
Build static graphnn v1.11 library
navigate to code/graphnn-1.11
modify configurations in make_common file

Build main nn code

navigate to code/nn

run test

navigate to code/nn

reproduce the results reported in paper

navigate to code/nn
modify the scripts under code/nn/scripts (or code/nn/server_scripts, the two folders 
have different parameter settings and path configurations)
execute the script

about data

all the synthetic datasets have been pushed to this repo. Some large datasets are not available here.