nanograv / 15yr_stochastic_analysis Goto Github PK

In the commands in the README file for the case where you are using miniconda, h5pulsar and ipympl need to be included. Even then, the kernel associated with my environment, did not appear automatically in the "Change Kernel" list in Jupyter. I don't know if this is a flaw in my environment setup or some other step is always needed. I was able to get Jupyter to see it by saying python -m ipykernel install --user --name=...

Figure 1, upper right appears to have some issue with the plot as can be seen here and is attached below. I haven't tracked down the exact cause of this yet, but I suspect it is an issue in the optimal_statistic_covariances.py file.

1. I think we should explain that (198.03261446315148, 45.16748075936892) returned by bfacs.log10_bf means that the Bayes factor in favor of model_3a is 198 +/- 45.
2. I think we should say that model_2a is "CURN", since we call it that in the paper.
3. Can you reproduce the 1e12 factor in favor of model_2a over intrinsic noise only ("IRN")? This would be useful to see because it cannot be done with product space sampling.