Open joshdempster opened 8 months ago
Hi, thanks for raising this issue and your suggestions. PR #35 contains a new check that gene_set_dict
is not empty. Agreed, ideally the format of gene_set_dict
is consistent. This should also be updated in the next release.
Trying to call
est_spectra
with a gene_set_dict of the form{"global": dict of gene sets}
anduse_cell_types=False
leads to an unbound local variable error foris_global
after training. Looking through the code, this happens because in this case,check_gene_set_dictionary
expects a single-layered dict forgene_set_dict
. This behavior isn't in the doc string. A couple suggestions:gene_set_dict
is consistent (my personal preference), or else havecheck_gene_set_dictionary
check ifglobal_key
is in the top layer ofgene_set_dict
before wrapping it in another layer.est_spectra
to be more explicit about the format and put an example of the format in the READMEgene_set_dict
is not empty after all changes are made to it and before training