Input array for fitting method

Question

This code returns the expected results. But there are 2 pandas methods involved. Can I use only 1 method or remove pandas from fit_transform?

from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
data = [-1, 2,1, 18]
scaler.fit_transform(pd.DataFrame(pd.Series(data)))

array([[0.        ],
       [0.15789474],
       [0.10526316],
       [1.        ]])

Update: I tried feeding the list, but got an error:

from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
data = [-1, 2,1, 18]
scaler.fit_transform(data)

# ValueError: Expected 2D array, got 1D array instead

This works, but getting wrong results:

from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
data = [-1, 2,1, 18]
scaler.fit_transform([data])

# array([[0., 0., 0., 0.]])

Perhaps reading the docs would have helped? It does say the function is meant to take lists, arrays, or similar. Not sure where you got the idea to introduce pandas objects here. — cs95
– cs95, Commented Dec 29, 2018 at 10:51

Courage · Accepted Answer · 2018-12-29 10:16:39Z

2

You can directly apply fit_transform like this without pandas

data = np.array(data).reshape(-1,1).astype(np.float32)
scaler.fit_transform(data)

answered Dec 29, 2018 at 10:16

Courage

7928 silver badges21 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Venkatachalam · Accepted Answer · 2018-12-29 16:40:28Z

1

Option 1:

You can directly create a dataframe from the list and feed it to the scaler, which would require any reshaping.

scaler.fit_transform(pd.DataFrame(data))

Option 2:

If you want to do with numpy,

scaler.fit_transform(np.array(data)[:,np.newaxis])

or you can go with the suggestion given by @Oswald

edited Dec 29, 2018 at 16:40

answered Dec 29, 2018 at 16:22

Venkatachalam

17k10 gold badges52 silver badges78 bronze badges

Collectives™ on Stack Overflow

Input array for fitting method

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related