Triangle-GB

Frequency-domain GB TDI-2.0 response, adapted from GBGPU (the GPU implementation of FastGB algorithm) to support the numerical orbit of Taiji and TDI-2.0. The responses are consistent with the time-domain simulations of Triangle-Simulator (see Example 0). Also offered is an illustrative example (Example 1) for the preliminary analysis of Taiji Data Challenge II, nevertheless, it should not be regarded as a solution to all the challenges. Especially, only individual GBs are analyzed in the example, while for realistic data, the major challenge we face would be the overlap of numerous signals.
Note that the future version of GBGPU will also support TDI-2.0 response function for generic orbits. Interested users are encouraged to follow its GitHub page for updates.

Installation

Tested platforms: Ubuntu22.04 (recommanded), MacOS15. While we donot ensure that installation will success on all the unix systems, even the aforementioned two.

1. Install Triangle-Simulator
   install Triangle-Simulator and activate the tri_env environment by

   conda activate tri_env 

2. Download or Clone the Repository, then

   cd Triangle-GB

3. Install Modified GBGPU

   More instructions on the installation, usage and troubleshooting of GBGPU can be found in the documentations.

• on linux:

  cd GBGPU_numorbit
  conda install -c conda-forge gcc_linux-64 gxx_linux-64 gsl Cython

  to get GPU support (optional), install cupy by pip (replace 92 by your own cuda toolkit version):

  pip install cupy-cuda92

  (Note that this could be complicated if you do not have CUDA installed in your system. Please follow the official guide of cupy if any problem occurs.)

  install package from source:

  python setup.py install

• on macos (arm chip):

  cd GBGPU_numorbit
  conda install -c conda-forge clang_osx-arm64 clangxx_osx-arm64 gsl Cython

  install package from source (there is currently no GPU support on macos):

  pip install . 
  

4. Install MCMC Tools to Run the Tutorial

   pip install eryn corner

Comparison with time-domain simulation

References

Please make sure to cite FastGB and GBGPU if Triangle-GB is used in your published research.

• Taiji Data Challenge II: the TDC II paper.
• The fast frequency-domain detector response of galactic binaries: N. J. Cornish et al, Phys.Rev.D76:083006,2007
• GBGPU: GBGPU documentations