Skip to content

Conversation

@jeremykubica
Copy link
Collaborator

Improve the efficiency of the ObsTableUniformRADECSampler by only testing the newly generated points (instead of checking everything, including the previously valid points). This does not fix the problems we occasionally see with the samples being out of bounds. But it makes debugging more efficient.

Also increase the coverage of the tests.

@jeremykubica jeremykubica requested a review from mi-dai September 30, 2025 20:17
@github-actions
Copy link

Before [ae17725] After [576a549] Ratio Benchmark (Parameter)
46.5±0.3μs 48.6±0.8μs 1.04 benchmarks.TimeSuite.time_make_new_salt3_model
29.1±0.3μs 30.1±0.5μs 1.04 benchmarks.TimeSuite.time_make_simple_linear_wavelength_model
672±10μs 695±20μs 1.03 benchmarks.TimeSuite.time_fnu_to_flam
9.36±0.3ms 9.55±0.3ms 1.02 benchmarks.TimeSuite.time_evaluate_salt3_model
4.80±0.1ms 4.89±0.09ms 1.02 benchmarks.TimeSuite.time_lightcurve_source
12.8±0.09μs 13.1±0.2μs 1.02 benchmarks.TimeSuite.time_sample_x1_from_hostmass
19.1±0.2ms 19.2±0.3ms 1.01 benchmarks.TimeSuite.time_chained_evaluate_sed
34.2±0.4ms 34.3±0.09ms 1 benchmarks.TimeSuite.time_additive_multi_model_source
554±9μs 556±9μs 1 benchmarks.TimeSuite.time_apply_passbands
7.58±0.1ms 7.55±0.05ms 1 benchmarks.TimeSuite.time_apply_white_noise

Click here to view all benchmarks.

@jeremykubica jeremykubica merged commit d55d637 into main Oct 1, 2025
7 checks passed
@jeremykubica jeremykubica deleted the ra_dec_sampler_optimization branch October 1, 2025 15:08
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants