See issue #353
FOR current_parallel IN SLICE(target_file.parallels_randomized, 0, 2500) only works for larger datasets. If the first 2500 matches in the randomized dataset are mostly filtered out and the result is only two items, the result is not correct.
Limiting it to 2500 matches AFTER filtering is better.
See issue #353
FOR current_parallel IN SLICE(target_file.parallels_randomized, 0, 2500)only works for larger datasets. If the first 2500 matches in the randomized dataset are mostly filtered out and the result is only two items, the result is not correct. Limiting it to 2500 matches AFTER filtering is better.