master: Refactoring of multilevelannotation step 5

RECETOX / recetox-xMSannotator

This is a custom adaptation of the original version of xMSannotator. It is a complete rewrite of the original functionality, following the same program structure.

GNU General Public License v3.0

5 stars 5 forks source link

master: Refactoring of multilevelannotation step 5 #63

Closed maximskorik closed 2 years ago

maximskorik commented 2 years ago

Refactored multilevelannotationstep5:
- removed unused variables & parameters
- removed data processing after writing data to disk
- simplified score recomputing and made it more descriptive
test-multilevelannotation5 is passing, but test-advanced_annotation is not. This likely has to do with #62.
Closes #60

hechth commented 2 years ago

Regarding the cluster and parallelism, did you remove it now or keep it as the sequential version?

I'd also suggest adding what we discussed yesterday to the documentation.

maximskorik commented 2 years ago

Regarding the cluster and parallelism, did you remove it now or keep it as the sequential version?

I'd also suggest adding what we discussed yesterday to the documentation.

I haven't pushed any changes to that part yet. I'm looking into it more precisely at the moment. On my machine sequential run causes all step 5 test-cases to fail with the only reason being different chemical scores (lower scores than expected).

hechth commented 2 years ago

How do you mean sequential? %do% or %dopar% without registering the cluster?

maximskorik commented 2 years ago

How do you mean sequential? %do% or %dopar% without registering the cluster?

Well, actually %do% without registered clusters causes the tests to fail as described above, while %dopar% without registered clusters passes the tests. This is weird as from my understanding %dopar% without clusters should function the same as %do%.

maximskorik commented 2 years ago

@hechth, maybe 20cce5a would be just enough? Essentially, we remove clusters that we suspect aren't used anyway while preserving the original functionality of the code. If I remember correctly, we are trying to preserve the functionality even if it includes some minor flaws, and this change does exactly that.