PURIFY
Next-generation radio interferometric imaging
|
PURIFY is an open-source collection of routines written in C++
available under the license below. It implements different tools and high-level to perform radio interferometric imaging, i.e. to recover images from the Fourier measurements taken by radio interferometric telescopes.
PURIFY leverages recent developments in the field of compressive sensing and convex optimization. Low-level functionality to solve the resulting convex optimisation is factored into the open-source companion code, SOPT, also written by the authors of PURIFY. For further background please see the reference section.
This documentation outlines the necessary and optional dependencies upon which PURIFY should be built, before describing installation, testing and usage details. Contributors, references and license information then follows.
PURIFY is written in C++11
. Required software and libraries, and their minimum supported versions, are listed below. The build system will attempt to automatically download and build the automatically included libraries. (an internet connection is required for this). Most dependencies are handled by the conan
package manager.
C++
dependencies:
In order to build PURIFY, you should have the following installed on your system.
C/C++
package manager. NOTE Conan 2.0 and later are not supported.C++
The build system of PURIFY will attempt to download and build these additional dependencies, depending on the build options passed to conan
. Most of them are automatically handled by conan
.
CMake
recipes. Downloaded automatically if absent.C++
linear algebra. Downloaded automatically by conan.C
and Fortran
subroutines for reading and writing data files in FITS (Flexible Image Transport System) data format. Downloaded automatically by conan.C++
. Downloaded automatically by conan.C++
unit-testing framework only needed for testing. Downloaded automatically by conan.C++
micro-benchmarking framework only needed for benchmarks. Downloaded automatically by conan.Using Conan v2 (recommended)
Conan is a C++ package manager that helps deal with most of the C++ dependencies as well as the PURIFY installation:
git clone
from the GitHub repository:conan
package for sopt
If you get an error about broken symlinks you can set skip_broken_symlinks_check = True
in your ~/.conan/conan.conf
file or set an environment variable
conan
:You can turn the various options on and off by adding flags to the conan install
command, e.g.
The full list of build options can be found in the conanfile. To install in directory INSTALL_FOLDER
, add the following options to the conan build command:
Using CMake
If the dependencies are already available on your system, you can also install PURIFY manually like so
On MacOS, you can also install most of the dependencies with Homebrew e.g.
The SOPT library includes an interface to TensorFlow for using trained models as priors in the Forward-Backward optimization algorithm. To build PURIFY with TensorFlow capability, some extra steps are currently required. We aim to simplify the build process in a future release.
cppflow
and create a conan
package usingFollow the nominal build instructions, making sure you enable the cppflow
option when building SOPT ...
```bash conan create /path/to/purify/sopt/ –build missing -o cppflow=on ```
... and PURIFY:
``` bash cd /path/to/purify mkdir build cd build conan install .. –build missing -o cppflow=on conan build .. ```
To check everything went all right, run the test suite:
The main purify
executable lives either in the build directory or in the in the bin
subdirectory of the installation directory. purify
has one required argument, it a string for the file path of the config file containing the settings.
purify path/to/config.yaml
.
A template with a description of the settings is included in the data/config
directory. When purify
runs a directory will be created, and the output images will be saved and time-stamped. Additionally, a config file with the settings used will be saved and time-stamped, helping for reproducibility and book-keeping.
A Dockerfile is available on DockerHub. We are currently not maintaining it, and cannot guarantee it is up to date. Use the below documentation at your own risk.
If you want to use Docker instead, you can build an image using the Dockerfile available in the repository or pulling it from DockerHub.
or
Then to use it, you should mount the directory with your data and config files to /mydata
in the container. To run the container and mount the directory is with:
That will start a shell inside the container in the /mydata
directory where you can see all the files from your /full/path/to/data
. There you can run purify
as shown above.`
Check the contributors page (github).
If you use PURIFY for work that results in publication, please reference the webpage and our related academic papers:
PURIFY Copyright (C) 2013-2019
This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details (LICENSE.txt).
You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
For any questions or comments, feel free to contact Jason McEwen, or add an issue to the issue tracker.
The code is given for educational purpose. The code is in beta and still under development.