readme.md

How to normalize data for machine learning and othe purposes

THe code

Importing libraries

import numpy as np
import pandas as pd
from sklearn.preprocessing import normalize,MinMaxScaler,RobustScaler,StandardScaler
import matplotlib.pyplot as plt

Creating dataset

#create dataset
data1 = np.random.normal(5,2,100)
data2 = np.random.randint(7,size=(100))
data2[53] = 25
data2[20] = 17
# x = np.linspace(data1.mean(), data1.max(), 1000)

The main normalization code

fig , ax = plt.subplots(2,5)
fig.set_size_inches(12,7)
fig.suptitle('Different types of normalization for data')
ax[0,0].hist(data1,color = "skyblue")
ax[0,0].set_title('Original data')
ax[0,0].set_ylabel('Normal distribution data')
ax[1,0].hist(data2,color = "skyblue")
ax[1,0].set_ylabel('Random data')
# normalize methode
data1Norm = normalize([data1])
data2Norm = normalize([data2])
ax[0,1].hist(data1Norm[0],color = '#32a852')
ax[0,1].set_title('normalize')
ax[0,1].set_xlim(0,1)
ax[1,1].hist(data2Norm[0],color = '#32a852')
ax[1,1].set_xlim(0,1)
# MinMaxScaler methode
scaler = MinMaxScaler()
data1Norm = scaler.fit_transform(data1.reshape(-1,1))
data2Norm = scaler.fit_transform(data2.reshape(-1,1))
ax[0,2].hist(data1Norm,color = '#eddf3e')
ax[0,2].set_xlim(0,1)
ax[0,2].set_title('MinMaxScaler')
ax[1,2].hist(data2Norm,color = '#eddf3e')
ax[1,2].set_xlim(0,1)

# RobustScaler methode
scaler = RobustScaler()
data1Norm = scaler.fit_transform(data1.reshape(-1,1))
data2Norm = scaler.fit_transform(data2.reshape(-1,1))
ax[0,3].hist(data1Norm,color = '#d13232')
ax[0,3].set_title('RobustScaler')
ax[0,3].set_xlim(-5,5)
ax[1,3].hist(data2Norm,color = '#d13232')
ax[1,3].set_xlim(-5,5)
# StandardScaler methode
scaler = StandardScaler()
data1Norm = scaler.fit_transform(data1.reshape(-1,1))
data2Norm = scaler.fit_transform(data2.reshape(-1,1))
ax[0,4].hist(data1Norm,color = '#8f32d1')
ax[0,4].set_xlim(-5,5)
ax[0,4].set_title('StandardScaler')
ax[1,4].hist(data2Norm,color = '#8f32d1')
ax[1,4].set_xlim(-5,5)

fig.savefig("norm.png")
plt.show()

The result