Build and use note

1. Download dependence:
libtorch and add libtorch to your environment path:
https://pytorch.org/cppdocs/installing.html

wget https://download.pytorch.org/libtorch/nightly/cpu/libtorch-shared-with-deps-latest.zip
unzip libtorch-shared-with-deps-latest.zip
export Torch_DIR=</absolute/path/to/libtorch>
echo "export Torch_DIR=</absolute/path/to/libtorch>" >> ~/.bashrc

clang-format is not reqiured, but if you want to use "make format", you need to have it installed

sudo apt install clang-format

2. build our project:
First way(using Makefile):
make all 
or 
make build
or 
make debug


Second way(by hand):
mkdir build
cd build
cmake  ..
cmake --build . --config Release
or
cmake --build . --config Debug

2.1 build on USC CARC 
module load gcc/8.3.0 
module load openmpi/4.0.2
module load cmake
export LD_PRELOAD=/spack/apps/gcc/8.3.0/lib64/libstdc++.so.6

build:
cmake  -DCMAKE_C_COMPILER=gcc   -DCMAKE_CXX_COMPILER=g++   ..


3. running:
mpirun -n 4 ./cpp-rl-training <path/to/config/file> > out   
or
mpirun -n 4 ./cpp-rl-training  > out   
in which 4 is the amount of nodes you want to use 
if config file is not specified, the default config file will be used (../config)
(> out) means everything which are supposed to be printed in terminal is instead printed in a file called out
  
3.1 running on CARC
use .sl file