Bugfix and enhancement in heterodyned likelihood

[thomasckng]

No description provided.

[thomasckng]

@coderabbitai review

[coderabbitai]

✅ Actions performed

Review triggered.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

[SSL32081]

This PR removes the evaluate_original function from Heterodyne likelihood, and calls the evaluate function from super class instead.

[SSL32081]

@coderabbitai review

[coderabbitai]

✅ Actions performed

Review triggered.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

[kazewong]

@thomasckng I think this is addressed in #237, can you double check if there is anything remained in this PR that is not included? Otherwise we can close this PR

[thomasckng]

@SSL32081 I have resolved the conflicts. Can you take a look at the recent commits and see if I missed anything?

[thomasckng]

@SSL32081 I have solved the slow __init__() issue. Can you take a look at 859d97f?

[Copilot]

Pull Request Overview

This PR addresses bugfixes and enhancements in the heterodyned likelihood implementation. The changes focus on improving code correctness, performance optimization through vectorization, and code maintainability.

Key changes include:

• Fixed critical bugs in reference parameter handling and frequency variable usage
• Replaced inefficient loop-based coefficient computation with vectorized operations using JAX/NumPy broadcasting
• Improved code organization by moving frequency grid initialization to proper location

Comments suppressed due to low confidence (2)

src/jimgw/core/single_event/likelihood.py:492

• [nitpick] The variable name pol is not descriptive. Consider using polarization or pol_key to make the code more readable and self-documenting.

            jnp.array([jnp.abs(h_sky[pol]) for pol in h_sky.keys()]), axis=0

src/jimgw/core/single_event/likelihood.py:745

• [nitpick] Using _ to discard the returned rng_key is appropriate, but consider adding a comment explaining what value is being discarded for clarity.

        _, best_fit, log_prob = optimizer.optimize(

[thomasckng]

@kazewong This PR has been updated. The major changes are a bugfix at line 558 and a significant enhancement in compute_coefficients().