Skip to content
This repository has been archived by the owner on Jan 7, 2023. It is now read-only.

Latest commit

 

History

History
60 lines (55 loc) · 1.85 KB

detailedinstructions.md

File metadata and controls

60 lines (55 loc) · 1.85 KB
Raw

Install Instructions for Neurovectorizer on Ubuntu 18.04

Table of Contents:

  1. Installing Anaconda, CMAKE, and LLVM
  2. Install Neurovectorizer Dependencies
  3. Edit configure.sh
  4. RISELab Students Directory Setup on Clusters

Anaconda CMAKE LLVM:

Skip any dependency you already have.

wget https://repo.anaconda.com/archive/Anaconda3-2020.02-Linux-x86_64.sh
bash Anaconda3-2020.02-Linux-x86_64.sh  # RISELab Students make sure to install in /data/[username]/ when asked rather than HOME
wget https://github.com/Kitware/CMake/releases/download/v3.17.1/cmake-3.17.1-Linux-x86_64.tar.gz
tar xzf cmake-3.17.0.tar.gz
cd cmake-3.17.0
./bootstrap
make
make install
cd ..
git clone https://github.com/llvm/llvm-project.git
cd llvm-project
mkdir build
cd build
cmake -DLLVM_ENABLE_PROJECTS=clang -G "Unix Makefiles" ../llvm
make

Install Neurovectorizer Dependencies:

First make sure you're in your anaconda environment.

conda install -c conda-forge clang
conda install -c conda-forge tensorflow
pip install tensorflow
pip install ray
pip install ray[rllib]
pip install clang

Edit configure:

Note the following paths are specific to the RISELab directory setup on clusters, so make adjustments as necessary. Change the folllowing lines:

CLANG_PATH=/data/[username]/dependencies/llvm-project/build/lib/libclang.so
CLANG_BIN_PATH=/data/[username]/anaconda3/bin/clang

Directory Setup:

The default working directory /home/eecs/[username] is far slower than working inside of /data directory so this should be your working directory.

cd /data/
mkdir [your username]
cd [your username]
mkdir dependencies

You may want to change your $HOME to point to /data/[username] for ease of access in the future.