Data release supporting:
- Characterization of merging black holes with two precessing spins. Viola De Renzis, Davide Gerosa, Geraint Pratten, Patricia Schmidt, Matthew Mould. Physical Review D 106 (2022) 084040. arXiv:2207.00030.
You are welcome to use this dataset in your research. We kindly ask you to cite the paper above. If you want to cite this data release specifically, the DOI code is: .
Data need to be downloaded from the github release page. The total size is ~35GB. To download the entire dataset use our download_all.py
script.
We provide two types of data products, named *_result.json
and *_chipav.dat
.
- The
result.json
files are the raw bilby outputs, see here for instructions. Each of these files contains a nested set of dictionaries with all the information of the injection (true values, waveform arguments, priors, posteriors, log evidence etc...). - The
chipav.dat
files contain the posterior samples of the averaged .$\chi_p$ estimator. Samples for the heuristic definition of$\chi_{p}$ are provided by default and can be found inresult.json
.
The data provided refer to the results of the paper as follows:
-
Data to reproduce Fig. 2 and Fig. 3 of the paper are listed as
DL_XXXX
whereDL
stands for luminosity distance andXXXX
refers to the value in Mpc. -
The 100 injections used in Fig. 4 and Fig. 5 are listed as
uni_XX
whereXX
ranges from 00 to 99. Injections are sorted in in ascendig order of$\chi_{p}$ . The prefixuni
stands for uniform because of our targeted$\chi_{p}$ distribution. For all these runs, injection and recovery are done with the IMRPhenomXPHM waveform models using the standard uninformative priors (see the paper). -
In Fig.6, the red posteriors correspond to the injections labeled
uni_XX
from above withXX
=[25,98]. The green posteriors are performed using volumetric priors and can be reproduced using files starting withprior_XX
. The yellow posteriors are done at higher SNR value and are provided asprior_XX_SNR_YY
. -
In Fig. 7, the posteriors obtained injecting and recovering with IMRPhenomXPHM are the ones are those labeled
uni_XX
withXX
=[21,81]. The other runs obtained with different combination of IMRPhenomXPHM, IMRPhenomTPHM and NRSur7dq4 (shortened respectively as XP, TP and NR) can be reproduced using files namedsyst_XX_WINJ_WREC
whereWINJ
andWREC
indicate the waveform model used in injection and recovery, respectively.