search

KacperKubara / distython

Distance metrics which can handle mixed-type data and missing values
58 stars 3 forks source link

Question with HEOM.py #14

Closed XinyiJi1 closed 4 years ago

XinyiJi1 commented 4 years ago

Hi, I met some problems when I tried to use the HEOM library with both numeric and categorical data. I feel the problem is the following: # Get the normalization scheme for numerical variables if normalised == "std": self.range = 4* np.nanstd(X, axis = 0) else: self.range = np.nanmax(X, axis = 0) - np.nanmin(X, axis = 0) Seems like the above code can not work well with categorical data (especially for np.nanmin() function).

KacperKubara commented 4 years ago

Probably a duplicate of https://github.com/KacperKubara/distython/issues/5?

XinyiJi1 commented 4 years ago

Probably a duplicate of #5?

Oh! I found it! Thank you! :)