import pandas as pd
from sklearn import preprocessing
x = df.values #returns a numpy array
min_max_scaler = preprocessing.MinMaxScaler()
x_scaled = min_max_scaler.fit_transform(x)
df = pd.DataFrame(x_scaled)

# Normalize the whole dataset before modeling
from sklearn.preprocessing import StandardScaler
X = StandardScaler()\	.fit(X)\	.transform(X.astype(float))
# There are other normalization methods available like MinMaxScaler, Z-score etc
# Alternative approach
from sklearn.preprocessing import normalize
X = normalize(X.astype(float), norm="l1")
# Alternative approach
from sklearn.preprocessing import Normalizer
X = Normalizer()\	.fit(X)\	.transform(X.astype(float))
# Alternative approach with scipy : normalize to a standard deviation of 1
from scipy.cluster.vq import whiten
scaled_data = whiten(data) # Works with multi-dimensional data
# Normalizing without library
# Feature Scaling
df["feature_scaled"] = df["col"]/ (df["col"].max())
# Min-max Scaling
df["minmax_scaled"] = (df["col"] - df["col"].min()) / (df["col"].max() - df["col"].min())
# Z-score
df["z_scaled"] = (df["col"] - df["col"].mean()) / df["col"].std()

from sklearn import preprocessing
data = [[ 1., -1., 2.], [ 2., 0., 0.], [ 0., 1., -1.]]
normalize(data)
# produces
array([[ 0.40824829, -0.40824829, 0.81649658], [ 1. , 0. , 0. ], [ 0. , 0.70710678, -0.70710678]])

from sklearn import preprocessing
normalizer = preprocessing.Normalizer().fit(X_train)
X_train = normalizer.transform(X_train)
X_test = normalizer.transform(X_test)

# define a method to scale data, looping thru the columns, and passing a scaler
def scale_data(data, columns, scaler): for col in columns: data[col] = scaler.fit_transform(data[col].values.reshape(-1, 1)) return data