Add Jacobian to GeneralizedLorentz1D, SmoothBrokenPowerLaw in stingray.simulator.models

[ankitkhushwaha]

TODO:

i have added the jacabian for each parameter in both GeneralizedLorentz1D, SmoothBrokenPowerLaw function.

i have confirmed that jacobian for function with respect to each parameter is correct.

@matteobachetti

[ankitkhushwaha]

I think changes are ready for review.
@matteobachetti

[matteobachetti]

@ankitkhushwaha thanks for your PR. However, I haven't noticed any tests failing, there are no issues open about it as far as I can tell. Could you be clearer about what is failing, and what your changes do?

[ankitkhushwaha]

@matteobachetti
in the first commit. Function GeneralizedLorentz1D, SmoothBrokenPowerLaw was returning the jacobian and function output as tuple. that's why some test are falling.

these test are fixed now.

TODO:
I have added the Jacobian functionality for GeneralizedLorentz1D, SmoothBrokenPowerLaw which will compute the derivative of their function with respective to each parameter.

[codecov]

Codecov Report

Attention: Patch coverage is 27.36842% with 69 lines in your changes missing coverage. Please review.

Project coverage is 46.31%. Comparing base (a91c031) to head (4fc246e).

Files with missing lines	Patch %	Lines
stingray/simulator/models.py	27.36%	69 Missing ⚠️

❗ There is a different number of reports uploaded between BASE (a91c031) and HEAD (4fc246e). Click for more details.

HEAD has 22 uploads less than BASE

Flag	BASE (a91c031)	HEAD (4fc246e)
	23	1

Additional details and impacted files

@@             Coverage Diff             @@
##             main     #889       +/-   ##
===========================================
- Coverage   96.03%   46.31%   -49.73%     
===========================================
  Files          48       48               
  Lines        9770     9855       +85     
===========================================
- Hits         9383     4564     -4819     
- Misses        387     5291     +4904     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[ankitkhushwaha]

Hey @matteobachetti, mostly this pr is ready to be reviewed, could you review this.
more tests needs to added that i will add in upcoming commits.

[ankitkhushwaha]

under the hood fit_driv function is called when
new_model_no_deriv = fitter_no_deriv(model_no_deriv, x, data, estimate_jacobian=True)
is called also model(x) will executes evalate function.
i am wondering why codecov is falling.
will check it later on .