Search code examples
machine-learningdata-sciencesklearn-pandas

Why isn't the first code not working, whereas second does?


First code snippet:

imp.fit_transform(dataset['Age'].values.reshape(-1,1))
dataset['Age'] = dataset['Age'].values.reshape(-1,1)

Second code snippet:

imp.fit(dataset['Age'].values.reshape(-1, 1))
dataset['Age'] = imp.transform(dataset['Age'].values.reshape(-1, 1))

The first code snippet does not make any change to the age column of my dataset.


Solution

  • The first code does not change the Age column because you are not asking it to do it ;) you apply the fit_transform function but you do not use it to change the dataset.

    I don't know what is imp, but a more correct implementation of the first piece of code would certainly be something like :

    dataset['Age'] = imp.fit_transform(dataset['Age'].values.reshape(-1,1))