These instructions are provided since we found RCA to be rather complex to install. (Some of these instructions assume an Ubuntu Linux operating system.) The recommended approach makes use of a cloned conda environment to avoid undesired side-effects, such as downgrading of the SciPy version, on the user's main conda environment.
Assuming the main conda environment is py38, create a new environment for RCA:
conda create --name py38rca --clone py38 conda activate py38rca
Install some RCA dependencies
pip install modopt conda install pyqtgraph
Build and install Sparse2D, which is required by PySAP:
sudo apt install cmake libcfitsio-dev git clone https://github.com/CosmoStat/Sparse2D.git cd Sparse2 mkdir build && cd build cmake .. make make install sudo mv -i sparse2d /opt export PATH=$PATH:/opt/sparse2d/bin
Install PySAP:
pip install python-pysap
Clone RCA from github and install it:
git clone https://github.com/CosmoStat/rca.git cd rca python setup.py install
If the shell in which these instructions are executed is closed, the RCA enviroment must be re-initialized by:
export PATH=$PATH:/opt/sparse2d/bin conda activate py38rca
It is recommended that the command above that adds /opt/sparse2d/bin to the path should be added to the user's shell initialization file (e.g. .bashrc).