Closed espenhgn closed 7 months ago
I've introduced some error-prone code in https://github.com/comorment/mixer/blob/main/usecases/run_mixer.ipynb; as np.unique(...) returns a sorted list of unique elements, hence the fname/trait pairing in such for loops can be completely arbitrary:
np.unique(...)
# fit1 fit1_job_ids = dict() for fname, trait in zip(np.unique(traitfiles_1 + traitfiles_2), np.unique(traitnames_1 + traitnames_2)):
It'd be better to define traitnames/files to iterate over in a dictionary as key/value pairs that shouldn't be untangled.
I've introduced some error-prone code in https://github.com/comorment/mixer/blob/main/usecases/run_mixer.ipynb; as
np.unique(...)returns a sorted list of unique elements, hence the fname/trait pairing in such for loops can be completely arbitrary:It'd be better to define traitnames/files to iterate over in a dictionary as key/value pairs that shouldn't be untangled.