Skip to main content

Parallelized rotation and flipping INvariant Kohonen maps

Project description

Build Status Join the chat at ascl:1910.001 PyPI version

Parallelized rotation and flipping INvariant Kohonen maps (PINK)

SOM of radio-synthesis data taken from the Radio Galaxy Zoo


  • C++ with ISO 14 standard (GCC 6.1 and above)
  • CMake >= 3.0
  • CUDA >= 9.1 (highly recommended because of the speed-up)
  • (optional for C++ dependencies) or
  • doxygen 1.8.13 (optional for developer documentation) will install automatically the C++ dependencies (PyBind11 and google-test). Otherwise you can also install these libraries yourself.


We provide deb- and rpm-packages at

or you can install PINK from the sources:

make install

PyPI installation

PINK is also available as PyPi package which can be installed by

pip install astro-pink


To train a the self-organizing map (SOM) please execute

Pink --train <image-file> <result-file>

where image-file is the input file of images for the training and result-file is the output file for the trained SOM. All files are in binary mode described here.

To map an image to the trained SOM please execute

Pink --map <image-file> <result-file> <SOM-file>

where image-file is the input file of images for the mapping, SOM-file is the input file for the trained SOM, and result-file is the output file for the resulting heatmap.

Please use also the command Pink -h to get more informations about the usage and the options.

Python scripts

For conversion and visualization of images and SOM some python scripts are available.

  • Convert binary data file from PINK version 1 to 2
  • Visualize the mapping result
  • Visualize binary images file format
  • Visualize binary SOM file format
  • SOM training using the PINK Python interface


The input data for the SOM training are radio-synthesis images of Radio Galaxy Zoo containing 176750 images of the dimension 124x124. The SOM layout is hexagonal of the dimension 21x21 which has 331 neurons (see image above). The size of the neurons is 64x64. The accuracy for the rotational invariance is 1 degree and the flip invariance is used.

PINK 1 Pink 2
CPU-1 35373
CPU-1 + NVIDIA Tesla P40 3069 909
CPU-1 + 2x NVIDIA Tesla P40 2069 636
CPU-1 + 4x NVIDIA Tesla P40 1891 858
CPU-2 + NVIDIA RTX 2080 673
CPU-3 + NVIDIA GTX 750 Ti 7185

All times are in seconds.

  • CPU-1: Intel Gold 5118 (2 sockets, 12 physical cores per socket)
  • CPU-2: Intel Core i7-8700K (1 socket, 6 physical cores per socket)
  • CPU-3: Intel Core i7-4790K (1 socket, 4 physical cores per socket)


Kai Lars Polsterer, Fabian Gieseke, Christian Igel, Bernd Doser, and Nikos Gianniotis. Parallelized rotation and flipping INvariant Kohonen maps (PINK) on GPUs. 24th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN), pp. 405-410, 2016. pdf


Distributed under the GNU GPLv3 License. See accompanying file LICENSE or copy at

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

astro-pink-2.4.1.tar.gz (71.6 kB view hashes)

Uploaded source

Built Distributions

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page