Version 0.1.1

I have just released version 0.1.1, which you can find here.

Changes since version 0.1 include:

  • adding support for multidimension numerical integrals via the Cubature algorithm;
  • implementing model-indepdent NP contributions in B→Dℓν decays;
  • exporting the Model class to the Python interface;
  • getting rid of the templated LogLikelihoodBlock::MultivariateGaussian, and replacing it with a class that handels the parameter size at run time.
  • providing active support for building and using EOS on MacOS X;
  • updating the manual and completing the part on software documentation, with help by Daniel Greenwald;

Contributors of these changes are:

  • Frederik Beaujean
  • Ahmet Kokulu
  • Danny van Dyk

EOS support on MacOS X

Thanks to a gratuitous donation of an iMac computer by Prof. Leo van Hemmen, we can now actively support building and using EOS on Mac OS X. We thank Professor van Hemmen very much for his donation!

The test hardware in use is limited to Mac OS 10.11 (El Capitan). We would be happy to receive reports about EOS builds on more recent Mac OS versions. The installation on MacOS via the Homebrew package manager is documented in the manual. The minimal commands necessary to build the most recent development version are

brew tap eos/eos
brew install --HEAD eos

docker.io images with EOS's built-time dependencies pre-installed

In order to speed up build time of our TravisCI test cases, docker images with the EOS build-time dependencies pre-installed are available over at the docker hub in eoshep/build-essentials. These are used as for CI testing as of commit 21f61a28. Even though it is a small annoyance that docker does not permit three-letter organization names, hence the eoshep name.

You can pull these images as you would pull any docker image using either

docker pull eoshep/build-essential:xenial

or

docker pull eoshep/build-essential:bionic

The source files for building these images are kept here. Please feel free to contribute your own docker files for whatever architecture you build EOS on.

New release (lambdab-polarised)

While working on their publication EOS-2017-02, Tom Blake and Michal Kreps derived the full set of angular observables for the decay Λb → Λ ℓ+- in the case of a polarised Λb baryon. They find a total of 34 angular observables compared to the 10 observables in the unpolarised case (compare the theory paper EOS-2014-01 and the fit to LHCb data in EOS-2016-02). The additional 24 observables can be classified as:

  1. A set of 2 observables that are copies of 2 of the unpolarised ones, diluted by the overall polarisation.
  2. A set of 22 observables that are sensitive to the same combinations of Wilson coefficients as the unpolarised ones, but with a different dependence on the hadronic matrix elements.

They modified the existing EOS code such that now all angular observables can be predicted. Their modifications have already been pushed to the master branch, and a corresponding tag has been pushed as well. A big thank you to Tom and Michal from my side for their contributions to EOS!

DFG Emmy-Noether Junior Research Group

Since August 1st 2017, the continued development and maintenance of EOS is funded by the Deutsche Forschungsgemeinschaft (DFG) as part of a DFG Emmy-Noether Junior Research Group.