You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The OrdinationResults object indicates in its docstring that eigvals and proportion_explained are to be pd.Series but this is not verified. A consequence is that, while np.ndarray can be provided and is a reasonable data type to use for those attributes, the IO registry fails on write as it correctly assumes that these attributes are pd.Series. Example traceback is below.
One reasonable path forward would be to relax the doctring, and cast on instantiation of OrdinationResults. Or, to gracefully error if the expected information isn't provided.
It may be the case that additional verification would be appropriate if for example there are index assumptions with the eigvals attribute
The
OrdinationResults
object indicates in its docstring thateigvals
andproportion_explained
are to bepd.Series
but this is not verified. A consequence is that, whilenp.ndarray
can be provided and is a reasonable data type to use for those attributes, the IO registry fails on write as it correctly assumes that these attributes arepd.Series
. Example traceback is below.One reasonable path forward would be to relax the doctring, and cast on instantiation of
OrdinationResults
. Or, to gracefully error if the expected information isn't provided.It may be the case that additional verification would be appropriate if for example there are index assumptions with the
eigvals
attributeThe text was updated successfully, but these errors were encountered: