Software developed at the CosmoStat lab at CEA Paris-Saclay.

This package has an option to install Sparse2D as a backend (disabled by default).

The package can be installed with pip as follows.

python -m pip install .

To install the Sparse2D backend along with the package installation, run:

BUILD_SPARSE2D=ON python -m pip install .

Sparse2D requires the following dependencies.

• Armadillo
• BigMac (macOS only)
• CFITSIO
• CMake
• FFTW
• GSL
• HEALPix
• OpenMP
• Python

To install the package requirements on macOS, it is recommended to use Homebrew.

brew tap sfarrens/sf
brew install armadillo bigmac cfitsio fftw gsl healpix libomp

To install the package requirements on Ubuntu, it is recommended to use apt.

sudo apt-get update
sudo apt-get install -y libarmadillo-dev libcfitsio-dev cmake libfftw3-dev libgsl-dev libhealpix-cxx-dev pkg-config build-essential 

If you have Docker installed, you can pull the latest build of the CosmoStat image as follows:

docker pull ghcr.io/cosmostat/cosmostat:master

No further installation is required.

To run a container on data in your current working directory, simply run:

docker run -v ${PWD}:/workdir --rm ghcr.io/cosmostat/cosmostat:master

The reference to ${PWD} can be replaced by the path to any directory on your system.

Additionally, you can run a Sparse2D executables dirextly from the image. For example, to run a bspline wavelet transform on a FITS image called myfile.fits you would run:

docker run -v ${PWD}:/workdir --rm ghcr.io/cosmostat/cosmostat:master mr_transform -t 2 myfile.fits myoutput.mr

Tip: If you don't want to constantly write the full Docker run command you can create an alias e.g.:

alias cosmostat="docker run -v ${PWD}:/workdir --rm ghcr.io/cosmostat/cosmostat:master"

then you can simply run e.g.:

cosmostat mr_transform -h

You can also run a Jupyter notebook with a Docker container as the backend in order to use the cosmostat package.

docker run -p 8888:8888 -v ${PWD}:/workdir --rm ghcr.io/cosmostat/cosmostat:master notebook