Problem #22. Different from actual definition of Normalization
nithinraok opened this issue ยท 3 comments
nithinraok commented
Normalization theoretically means subtracting by mean and dividing by standard deviation so that overall data has zero mean and unit variance.
So the code becomes
Z=np.random.random((5,5));
a=(Z-np.mean(Z))/(np.std(Z))
print(a)
lexnekr commented
I think you're right about normalization.
https://en.wikipedia.org/wiki/Normalization_(statistics)
The task now uses the formula not "normalization", but "Feature scaling"
P.S. I created pull request.
nithinraok commented
Looks good to me
Yossarian0916 commented
a=(Z - Z.mean()) / Z.std()
should actually be called standardization and z-score,
a=(Z - Z.mean()) / (Z.max() - Z.min())
is the feature scaling and normalization