feat: add nonlinear timing model#240
Open
davecwright3 wants to merge 12 commits intonanograv:masterfrom
Open
Conversation
davecwright3
force-pushed
the
feat/nonlinear-timing
branch
2 times, most recently
from
fd0d103 to
f70023b
Compare
Author
|
I broke down the monolithic commits into smaller, more logical commits. Also rebased onto master to include flake8 change. |
davecwright3
force-pushed
the
feat/nonlinear-timing
branch
from
f70023b to
291b6ca
Compare
Author
|
GH actions tests are failing because SciPy likes to make breaking changes on minor version increases. We should either pin SciPy or change the methods we call from SciPy. |
davecwright3
changed the title
Adds new prior methods useful for nonlinear timing to timing.py Co-authored-by: Andrew Kaiser <ark0015@mix.wvu.edu>
Co-authored-by: Andrew Kaiser <ark0015@mix.wvu.edu>
Updates tm_delay and timing_block with nonlinear timing. Also adds dm_block and other supporting methods. Co-authored-by: Andrew Kaiser <ark0015@mix.wvu.edu>
Co-authored-by: Andrew Kaiser <ark0015@mix.wvu.edu>
Adds new functionality for nonlinear timing, including specific jump proposals and restrictions to pulsar mass. Co-authored-by: Andrew Kaiser <ark0015@mix.wvu.edu>
Co-authored-by: Andrew Kaiser <ark0015@mix.wvu.edu>
Adds nonlinear timing model to single pulsar noise model. Co-authored-by: Andrew Kaiser <ark0015@mix.wvu.edu>
Adds nonlinear timing model to model_general. Co-authored-by: Andrew Kaiser <ark0015@mix.wvu.edu>
Co-authored-by: Andrew Kaiser <ark0015@mix.wvu.edu>
Co-authored-by: Andrew Kaiser <ark0015@mix.wvu.edu>
Co-authored-by: Andrew Kaiser <ark0015@mix.wvu.edu>
SciPy has deprecated interp2d used in the 2D KDE method. Their suggested backwards compatible method is not actually backwards compatible because it doesn't expose the same arguments to users, like the fill value outside of the domain. The suggested new methods take different inputs, so it is unclear if they will work with the empirical distribution code. This needs to be further tested before changing interp2d.
davecwright3
force-pushed
the
feat/nonlinear-timing
branch
from
bdbe656 to
06bd8e4
Compare
Member
|
Hi @davecwright3 , If you make your PR depend on #248 then the tests will probably run |
Member
|
Hah, what I realize now: that particular PR I reference of course makes it so that potentially libstempo is not installed. And that's the whole point of the PR here. The irony... |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This is a rework of #187.
I went through the commit history and pulled out any changes that weren't linting, updates from dev, etc (via https://github.com/Wilfred/difftastic AST diffs). The code is passing all tests locally, and the example notebook is producing output matching what @Hazboun6 had in the original PR.
This PR also includes documentation to go along with the new features.