Hands-on
Scarlet Installation Instructions
Participants to the hands-on sessions are strongly encouraged to have a tested installation of Scarlet, and a preliminary knowledge of its main features.
Full documentation, including the install instructions, is available at
https://pmelchior.github.io/scarlet/index.html
We explicitly note that Scarlet is based on Python3.7, and that the following dependencies are needed
numpy
proxmin (proximal algorithms used to minimize the likelihood)
pybind11 (integrate C++ code into python)
peigen (used to load the Eigen headers if they are not already installed)
autograd (needed to calculate gradients during optimization)
The most efficient way to have all the needed modules, without interfering with your usual code, is to install a Conda distro, which can create a Scarlet-only environment.
(Astro)-(Mini)-Conda is available at
https://docs.conda.io/projects/conda/en/latest/index.html
Linux (Debian) users can also install Anaconda with the following procedure
curl https://repo.anaconda.com/pkgs/misc/gpgkeys/anaconda.asc | gpg --dearmor > conda.gpg
sudo install -o root -g root -m 644 conda.gpg /etc/apt/trusted.gpg.d/
sudo apt-get update
sudo apt-get install conda
source /opt/conda/etc/profile.d/conda.sh
Then modify .bashrc by inserting
export PATH=/opt/conda/bin:$PATH
Mac users may experience some troubles with some libraries and the gcc compiler. These problems can be overcome with the following command:
CC=clang python setup.py install
Finally, here is an example of Conda environment creation and activation (https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html#creating-an-environment-with-commands)
conda create -n scarlet scipy
conda activate scarlet
pip install proxmin
pip install astropy
pip install autograd
Data download
Additional data can be also downloaded from:
https://drive.google.com/drive/folders/1ui7snv4JlAIgvMTfJIEpiWWyInjOjQUJ?usp=sharing
and be copied in the data/ folder of your Scarlet installation.
We warmly thank Fred Moolekamp for having provided us the additional data.
