Add x0 functionality to mcmc sampling.

[arm61]

@Lorentyfle would you be able to have a look at this PR?

[arm61]

Merging this would close #214

[github-actions]

☂️ Python Coverage

current status: ✅

Overall Coverage

Lines	Covered	Coverage	Threshold	Status
2158	2049	95%	0%	🟢

New Files

No new covered files...

Modified Files

File	Coverage	Status
kinisi/fitting.py	96%	🟢
kinisi/tests/test_fitting.py	100%	🟢
TOTAL	98%	🟢

updated for commit: 36a005d by action🐍

[Copilot]

Pull request overview

This PR extends kinisi.fitting.FittingBase.mcmc to accept an optional x0 initial position for MCMC sampling, and adds tests to validate correct behavior and unit checking.

Changes:

• Add an optional x0 argument to FittingBase.mcmc with basic unit validation.
• Add unit tests covering x0 usage and incorrect-unit rejection.
• Introduce pytest usage in test_fitting.py for exception assertions.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
kinisi/fitting.py	Adds x0 support to MCMC initialization and updates docstring.
kinisi/tests/test_fitting.py	Adds tests for x0 initialization and wrong-unit handling.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Lorentyfle]

To review this pull request functionality, as github verified all the tests I decided to check if data are well reproduced using my own data and comparing it to result given by the old version of Kinisi:

Code used for testing this PR:

from scipy.stats import lognorm
###############################################################################
# BOUNDS
###############################################################################

x0 = (
    # activation_energy_low
        1.78 * sc.Unit("eV")
    ,
      # activation_energy_high
        1.05 * sc.Unit("eV")
    ,
    # preexponential_factor = D(T0)
        1.75e-5 * td_D.data.unit
    ,
      # Tc
        1380 * sc.Unit("K")
    )
priors = (
    # activation_energy_low
        lognorm(s=1.78,scale=0.2)
    ,
      # activation_energy_high
        lognorm(s=1.05,scale=0.2)
    ,
    # preexponential_factor = D(T0)
        lognorm(s=0.5,scale=1.75e-5)
    ,
      # Tc
        lognorm(s=1380,scale=100)
    )

###############################################################################
# MCMC
###############################################################################

print("===> Launching piece-wise arrhenius")

tdsA = PiecewiseArrhenius(
    td_D,
    priors=priors
)
tdsA.mcmc(x0=x0)
print("===> MCMC completed")

Old Kinisi results:

Current PR results:

Conclusion

Because the results are identical in both versions of Kinisi, the addition of the functionality is correct and fully functional.