These instructions are essentially those available at https://github.com/eth-sri/eran with additional comments on how to circumvent some hiccups that may arise.

We will use anaconda to install various tools required to get "eran" up and running.

Installing Anaconda:
	Either go to https://docs.anaconda.com/free/anaconda/install/linux/ and follow the detailed instructions to install anaconda. Alternatively, follow the instructions given below from a Linux shell.

	1. Type "wget https://repo.anaconda.com/archive/Anaconda3-2023.07-1-Linux-x86_64.sh"
	2. Type "bash Anaconda3-2023.07-1-Linux-x86_64.sh"
	3. Press "Enter" to review the license agreement. Then press and hold Enter to scroll
	4. Enter "yes" to agree to the license agreement.
	5. Use Enter to accept the default install location. 
	6. Enter "yes" to initialize Anaconda Distribution by running conda init
	7. Type "conda init" to activate conda package or run the command "source ~/anaconda3/bin/activate"
	
Instructions to install ERAN:	
	1. Create the anaconda envirement variable "eran" with python 3.6 by following commands:  
		conda create -n eran python=3.6
	2. Activate the envirement variable "eran" by following commands:
		conda activate eran
	3. Clone the ERAN from git by following commands: 
		git clone https://github.com/eth-sri/ERAN.git
		cd ERAN
	4. Install the basic linux packages from the following commands:
		sudo apt-get install m4
		sudo apt-get install build-essential
		sudo apt-get install autoconf
		sudo apt-get install libtool
		sudo apt-get install texlive-latex-base
		sudo snap install cmake --classic
 	5. Install gmp (Note that command "make install" in steps 5,6,7,and 10 may fail due to permission issue, in that case one has to run the command sudo make install): 
 		wget https://gmplib.org/download/gmp/gmp-6.1.2.tar.xz
		tar -xvf gmp-6.1.2.tar.xz
		cd gmp-6.1.2
		./configure --enable-cxx
		make
		make install
		cd ..
		rm gmp-6.1.2.tar.xz
	6. Install mpfr: 
		wget https://files.sri.inf.ethz.ch/eran/mpfr/mpfr-4.1.0.tar.xz
		tar -xvf mpfr-4.1.0.tar.xz
		cd mpfr-4.1.0
		./configure
		make
		make install
		cd ..
		rm mpfr-4.1.0.tar.xz
	7. Install cddlib: 
		wget https://github.com/cddlib/cddlib/releases/download/0.94m/cddlib-0.94m.tar.gz
		tar zxf cddlib-0.94m.tar.gz
		rm cddlib-0.94m.tar.gz
		cd cddlib-0.94m
		./configure
		make
		make install
		cd ..
	8. Install Gurobi:
		wget https://packages.gurobi.com/9.1/gurobi9.1.2_linux64.tar.gz
		tar -xvf gurobi9.1.2_linux64.tar.gz
		cd gurobi912/linux64/src/build
		sed -ie 's/^C++FLAGS =.*$/& -fPIC/' Makefile
		make
		cp libgurobi_c++.a ../../lib/
		cd ../../
		cp lib/libgurobi91.so /usr/local/lib	#if this command fails due to permission related issues, then one has to run with sudo privilege
		python3 setup.py install #if this command gives you permission related errors, then change the permission of gurobi912 directory recursively to local user. 
		cd ../../
	9. Update environment variables:
		export GUROBI_HOME="$PWD/gurobi912/linux64"
		export PATH="$PATH:${GUROBI_HOME}/bin"
		export CPATH="$CPATH:${GUROBI_HOME}/include"
		export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/lib:${GUROBI_HOME}/lib
	10. Install ELINA:
		git clone https://github.com/eth-sri/ELINA.git
		cd ELINA
		if you have gpu
			./configure -use-deeppoly -use-gurobi -use-fconv -use-cuda
			cd ./gpupoly/
			cmake .
			cd ..
		else #for cpu only
			./configure -use-deeppoly -use-gurobi -use-fconv
		make
		make install
		cd ..
	11. Install DeepG:
		git clone https://github.com/eth-sri/deepg.git
		cd deepg/code
		mkdir build
		make shared_object
		cp ./build/libgeometric.so /usr/lib	#If this command failed due to permission issue then one has to run with sudo previllage
		cd ../..
	12. pip3 install -r requirements.txt
	13. Get free academic license for gurobi from https://user.gurobi.com/download/licenses/free-academic.


One can download example networks to check local robustness properties from https://github.com/eth-sri/eran.  You need to download some of these networks explicitly in .onnx format before running eran.  They are not automatically downloaded when you clone the "eran" repository from github

Command to run the tool (in the eran environment created using anaconda):
	Change your directory to tf_verify.  Then run
	
	python3 . --netname downloaded_network_path --epsilon 0.1 --domain deeppoly --dataset mnist

Suppose my network path is: ~/Downloads/mnist_relu_3_50.onnx
	python3 . --netname ~/Downloads/mnist_relu_3_50.onnx --epsilon 0.1 --domain deeppoly --dataset mnist