A thing that people sometimes do is run the taxonomic assignment/collapse algorithms multiple times and see how the output differs. We could potentially automate this (in parallel even!) by allowing the user to pass a range (or list?) of values for each parameter and then executing the process(es) for each option. They already put their output in separate subdirectories so it wouldn't be too difficult to get this to work.
Possible ways of doing this:
a range:
--param start-end[,step]
Would execute the process for each possible value of param (from start to end), optionally stepping by the value of step, or defaulting to step by whatever the smallest non-zero place is (.e.g., 10-30 would step by 10, 1-10 would step by 1, 0.1-0.5 would step by 0.1, etc.)
a list (this would require the use of a YAML options file:
A thing that people sometimes do is run the taxonomic assignment/collapse algorithms multiple times and see how the output differs. We could potentially automate this (in parallel even!) by allowing the user to pass a range (or list?) of values for each parameter and then executing the process(es) for each option. They already put their output in separate subdirectories so it wouldn't be too difficult to get this to work.
Possible ways of doing this:
a range:
Would execute the process for each possible value of
param(fromstarttoend), optionally stepping by the value ofstep, or defaulting to step by whatever the smallest non-zero place is (.e.g., 10-30 would step by 10, 1-10 would step by 1, 0.1-0.5 would step by 0.1, etc.)a list (this would require the use of a YAML options file: