Install HyperMapper

Dependencies:

In general, the use of Conda is recommended because it provides user-friendly access to a wealth of packages from the scientific Python ecosystem. Conda is a good default choice in Linux and OSX. Tested and preferred install anaconda3-4.8.3, other versions of anaconda3 should just work.

anaconda3-4.8.3 (for Python 3, Numpy, Matplotlib, etc.).

You can directly install Anaconda from here. For Linux, you can install with:

wget https://repo.anaconda.com/archive/Anaconda3-2020.02-Linux-x86_64.sh
bash Anaconda3-2020.02-Linux-x86_64.sh

Then reload the terminal for the changes to take effect.

Install:

git clone https://github.com/luinardi/hypermapper.git
cd hypermapper
export HYPERMAPPER_HOME=/path/to/hypermapper

Note that:

if HYPERMAPPER_HOME is not set up then "." will be used. HYPERMAPPER_HOME is useful when calling HyperMapper from another software like it is usual in Client-Server mode.

Using HyperMapper

Add the scripts directory to your PYTHONPATH to be able to import HyperMapper's modules. This can be done in your terminal with:

export PYTHONPATH=$PYTHONPATH:/path/to/hypermapper/scripts

Or in Python during runtime with:

import sys
sys.path.append('scripts')

Note that updating the PYTHONPATH during runtime will cause the changes to be reverted after execution.

Optional Dependencies:

These dependencies are not needed unless you are an advanced user.

pygmo This dependence is only needed for the plot_hvi.py script that enables to compare two Paretos using the HVI metric, see here for more info. You can find more info here: https://esa.github.io/pagmo2/install.html. WARNING: the python package pygmo crashes with anaconda3-5.0.0 while it works for anaconda3-4.4.0, this bug is unreported on the web and I found it after hours of exploration. anaconda3-5.0.0 only worked once, that was on Ubuntu 14.04.5 LTS.

Quick install on Linux and OSX:
```
conda config --add channels conda-forge
conda install pygmo
```
You can check that pygmo is installed by typing:
```
conda list | grep -i pygmo
```
Output should be something like:
```
pygmo     2.6     np113py36_1     conda-forge
```
Then you can check the import in python, if the import fails it is just because you have multiple versions of python and you are using the wrong version, i.e. the one that doesn't have pygmo installed, or because you are using a version of anaconda that crashes when loading the module, e.g. anaconda3-5.0.0:
```
python
from pygmo import *
```