Minor adjust to the ROQ weight generation

1. Convert the light travel time explicitly calculated value: uses 1.1 * R_earth / c (i.e. 10% more than the expected maximum)
2. Use a single fixed time_space: previously, the time resolution was set, then recalculated from the difference in the first two elements of time_samples. This could lead to buggy behaviour since that is no longer an evenly spaced array after the subtraction from the start time is made.
3. Add info statements
4. Adds a check if the seglen is shorter than the duration
5. Apply a safety factor of 5 to ensure the time step is short enough. This will increase the size of the ROQ array basis by ~ a factor of 5, but in testing I found it helped remove bias from examples just searching over RA, DEC, and geocent_time.
6. Add explicit clean up of memory

Closes #372 (closed)

added 1 commit

• c2dedc78 - Fix flake8 issue

Compare with previous version

~~Running the review script here to test this: https://ldas-jobs.ligo.caltech.edu/~gregory.ashton/bilby_tests/review/outdir_fiducial_bbh_128s_190521_2254_ROQ/~~

changed the description

added 1 commit

• 1ee18f8b - Force memory cleaning

Compare with previous version

changed the description

added 1 commit

• fb0861f7 - Move gc to top-level import

Compare with previous version

mentioned in issue #372 (closed)

added 1 commit

• 8c8c5454 - Simplify the shifted data calculation and and use the assume sorted

Compare with previous version

added 1 commit

• b4289876 - Fix type in the ROQ likelihood

Compare with previous version

resolved all discussions

added 1 commit

• 3b7cb586 - Revert to previous method of calculating roq frequencies

Compare with previous version

resolved all discussions

approved this merge request

added 1 commit

• 544155fe - Extend the light crossing time to the full radius and add warning message

Compare with previous version

added 1 commit

• b3077544 - Reduce to 10% extra delay time

Compare with previous version

changed milestone to %0.5.1

changed the description

128s basis review test run available here: https://ldas-jobs.ligo.caltech.edu/~gregory.ashton/bilby_tests/review/outdir_fiducial_bbh_128s_ROQ/result/fiducial_bbh_128s_ROQ_190525_1916_combined_corner.png

Note, also uses the dynesty fix for periodic boundaries

approved this merge request

approved this merge request

merged

mentioned in commit f9379f4f