정규화

최대값으로 정규화

import pandas as pd
import numpy as np

df = pd.read_csv('auto-mpg.csv')

df.columns = ['mpg', 'cylinders',
              'displacement', 'horsepower',
              'weight', 'accelerationi',
              'model year', 'origin', 'name']

df['horsepower'].replace('?', np.nan, inplace=True)
df.dropna(subset=['horsepower'], axis=0, how='any', inplace=True)
df['horsepower'] = df['horsepower'].astype('float')

df['horsepower'] = df['horsepower']/df['horsepower'].max()
print(df['horsepower'].head())

0    0.717391
1    0.652174
2    0.652174
3    0.608696
4    0.860870
Name: horsepower, dtype: float64

최대 - 최소 방법

$$ \frac{N - \min}{\max - \min} $$

각 원소를 위와 같은 방식으로 정규화 할 수 있다.

import pandas as pd
import numpy as np

df = pd.read_csv('auto-mpg.csv')

df.columns = ['mpg', 'cylinders',
              'displacement', 'horsepower',
              'weight', 'accelerationi',
              'model year', 'origin', 'name']

df['horsepower'].replace('?', np.nan, inplace=True)
df.dropna(subset=['horsepower'], axis=0, how='any', inplace=True)
df['horsepower'] = df['horsepower'].astype('float')

min_x = df['horsepower'] - df['horsepower'].min()
min_max = df['horsepower'].max() - df['horsepower'].min()
df['horsepower'] = min_x / min_max

print(df['horsepower'].head())

0    0.646739
1    0.565217
2    0.565217
3    0.510870
4    0.826087
Name: horsepower, dtype: float64

back_to_top