Resolve "parameterized eos sampling"
Merge request to include the spectral decomposition parameterized equation of state (EOS) simulation code and ability to sample in EOS model parameters.
Merge request reports
Activity
added 1 commit
- 87855000 - Added ability to sample in Spectral EOS parameters
added 2 commits
added 1 commit
- 78b9717d - Added option to use linear interpolation scheme in EOS functions.
added 3 commits
- b7496065 - Style guide fixes
- 8539c225 - Merge branch '369-eos-sampling' of https://git.ligo.org/matthew.carney/bilby into 369-eos-sampling
- 37a9934f - Fixed PEP8 style errors
added 1 commit
- c19342ee - Fixed optional dependency error in eos module.
added 93 commits
-
7325c30a...6415c788 - 91 commits from branch
lscsoft:master - d6917038 - Fixed merge conflict with master
- 9fbfb671 - Moved piecewise polytrope model over to another branch. This branch only...
-
7325c30a...6415c788 - 91 commits from branch
added 1 commit
- cffcd25b - Standardized EOSFamily method naming convention.
Hi all,
I've removed the WIP tag from this merge request as I think it's ready to be reviewed and merged in. This merge request only includes the spectral decomposition EOS model but I'll add in the piecewise polytrope model and potentially some others in the future. I'd prefer to keep those as separate features to add instead of adding them on to this MR. Please let me know if you'd like me to change anything.
added 1 commit
- 09608472 - Fixed units issue in plotting code and numerical integration relative error threshold.
added 1 commit
- 905536bb - Fixed unit issue with EOSFamily lambda_of_m method.
added 1 commit
- aac10272 - added np.nan_to_num() to causality constraint.
added 1 commit
- 00da63fe - Changed -np.inf to np.nan_to_num(-np.inf) in EOS causality check.
added 1 commit
- 8807b1b8 - Changed -np.inf to np.nan_to_num(-np.inf) in EOS causality check.
-
@matthew.carney generally speaking, this is in good shape. I left a bunch of comments, mostly focussed on style things as I'm not too sure how to review that the code is behaving as it should without unit testing.
Generally speaking, there need to be better documentation (and correspondingly fewer inline comments), more explicit variable names, and maybe shorter methods to break up the longer ones.
The things I couldn't comment on before are:
- you appear to have added a few random files, e.g.,
.swpfiles. - it would be nice to better document the EOS files. Maybe that can be in an accompanying README?
- it would be great to get a page in the docs, that would involve creating a new file in the
docsfolder to explain the features and usage. - we also need some kind of unit tests on this. Are there analytic tests that can be done for the different EOS models?
- you appear to have added a few random files, e.g.,
added 1 commit
- dbd957a0 - Added check to ensure proposed mass point is below maximum allowed mass of EOS.
added 39 commits
-
dbd957a0...6e1fec17 - 38 commits from branch
lscsoft:master - 8f796d43 - Merged latest version of master, bilby 0.5.7
-
dbd957a0...6e1fec17 - 38 commits from branch
added 1 commit
- 2bf7dfba - Adiabatic check is now imposed when sampling. This fixes an issue where too...
changed milestone to %0.5.9
@matthew.carney, I'd like to get this merged in to avoid it drifting too much. What sort of timescale do you think you'll be able to address the comments above? At the same time could you rebase it to the current master to make sure there aren't any conflicts.
-
@matthew.carney I'm moving this to 0.6.0
