Can scikit learn fit_transform and convert my original framework?

Question

I use scikit learning StandardScaler () and notice that after applying transform (xtrain) or fit_transform (xtrain) it also changes my xtrain dataframe. Is this supposed to happen? How can I avoid changing the StandardScaler of my framework? (I tried using copy = False)

xtrain.describe()    #everything ok here
scalar = StandardScaler()
xtrain2 = scalar.fit_transform(xtrain)   

At this point, I expected xtrain to NOT change, and xtrain2 to become a scaled version of xtrain. But when I start to describe () on 2 frames of data, I see that they are both the same and both were scaled. Why is this?

I experience the same problem when I do this:

scalekey = scalar.fit(xtrain)
xtrain2 = scalekey.transform(xtrain)

Answer 1

You can take a copy and pass this in order not to change your df:

xtrain2 = xtrain.copy()
scalar.fit_transform(xtrain2) 

docs , StandardScaler - copy=True, df.

Answer 2

0,16x pandas, scikit-learn. .