Z-score Standardization in Python

Z-score 데이터 표준화 방법에 대해서 살펴보자.

Sample Dataset

import pandas as pd
from sklearn.datasets import load_iris

iris = load_iris()
df = pd.DataFrame(data=iris.data, columns=iris.feature_names)
df['target'] = iris.target

Method 1 : Pandas

$z-score = (X - \mu) \div \sigma$

for col in df.select_dtypes('float').columns:
  df[col + '_zscore'] = (df[col] - df[col].mean()) / df[col].std(ddof=0)

Method 2 : scipy.stats.zscore

• https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.mstats.zscore.html

from scipy.stats import zscore
zscore(df.select_dtypes('float'))
# or,
pd.DataFrame(
    zscore(df.select_dtypes('float')),
    columns=df.select_dtypes('float').columns
)
# or,
df.select_dtypes('float').apply(zscore)

Method 3 : StandardScaler

from sklearn.preprocessing import StandardScaler

scaler = StandardScaler()
pd.DataFrame(
    scaler.fit_transform(df.select_dtypes('float')),
    columns=df.select_dtypes('float').columns
)