Gaussian Processes Tool

[Gaurav17Joshi]

Gaussian Processes Feature for QPO detection

Addressing #660 . This is an implementation for the GP tool (mainly for detection and parameter analysis for QPOs) . This addition makes use of tinygp library for Gaussian Processes implementation, and jaxns for nested sampling. The new feature is in the stingray.modeling.gpmodelling.py file. It has three main features:-

1. get_prior: This function makes the prior function for a specified prior dictionary.
2. get_likelihood: This function makes the log_likelihood function for the given Kernel, Mean model and lightcurve.
3. GPResult class: The class which takes a Lightcurve, and performs Nested Sampling for a given prior and likelihood.

Work Remaining To be done:-

• Add Object Oriented plotting to GPResult.
• Potentially change the name of file to GP (from GP modelling)
• Add the Demonstration notebook to Stingray Notebooks.
• Improve the documentation

[codecov[bot]]

Codecov Report

Merging #739 (e9ef4df) into main (5d633c1) will increase coverage by 0.16%. Report is 76 commits behind head on main. The diff coverage is 98.21%.

@@            Coverage Diff             @@
##             main     #739      +/-   ##
==========================================
+ Coverage   97.13%   97.29%   +0.16%     
==========================================
  Files          42       43       +1     
  Lines        7929     7918      -11     
==========================================
+ Hits         7702     7704       +2     
+ Misses        227      214      -13     

Files	Coverage Δ
stingray/__init__.py	100.00% <100.00%> (ø)
stingray/crossspectrum.py	99.00% <100.00%> (+0.78%)	:arrow_up:
stingray/events.py	99.53% <100.00%> (-0.47%)	:arrow_down:
stingray/fourier.py	99.79% <100.00%> (+0.05%)	:arrow_up:
stingray/lightcurve.py	97.89% <100.00%> (+<0.01%)	:arrow_up:
stingray/lombscargle.py	100.00% <100.00%> (ø)
stingray/powerspectrum.py	99.67% <100.00%> (+0.22%)	:arrow_up:
stingray/pulse/overlapandsave/test_ols.py	100.00% <100.00%> (ø)
stingray/utils.py	98.85% <ø> (+0.35%)	:arrow_up:
stingray/modeling/gpmodeling.py	95.78% <95.78%> (ø)

:mega: We’re building smart automated test selection to slash your CI/CD build times. Learn more

[matteobachetti]

Additional request: please add a short sentence in the installation instructions, about the added dependencies in case one wants to use the GP tool

[pep8speaks]

Hello @Gaurav17Joshi! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

• In the file stingray/modeling/gpmodeling.py:

Line 40:50: W291 trailing whitespace Line 622:101: E501 line too long (109 > 100 characters)

• In the file stingray/modeling/tests/test_gpmodeling.py:

Line 25:1: E402 module level import not at top of file Line 26:1: E402 module level import not at top of file Line 27:1: E402 module level import not at top of file

[dhuppenkothen]

The black issue is my fault, I will fix this in the next PR. Other than that, all the tests are passing, all of Matteo's and my nitpicky comments have been addressed, and I think this is ready for merging.

Thank you, @Gaurav17Joshi for your fantastic work! It was a big effort, but I'm so excited that this method is now in stingray!