Ex:05 Feature Generation

AIM:

To read the given data and perform Feature Generation process and save the data to a file.

Explanation:

Feature Generation (also known as feature construction, feature extraction or feature engineering) is the process of transforming features into new features that better relate to the target.

Algorithm:

Step 1: Read the given Data.

Step 2: Clean the Data Set using Data Cleaning Process.

Step 3: Apply Feature Generation techniques to all the feature of the data set.

Step 4: Save the data to the file.

Name: Karthick P
Register Number: 212222100021

Program

Encoding data.csv

import pandas as pd
df=pd.read_csv('Encoding Data.csv')
df.head()
df['ord_2'].unique()
from sklearn.preprocessing import LabelEncoder,OrdinalEncoder
climate = ['Cold','Warm','Hot']
en= OrdinalEncoder(categories = [climate])
df['ord_2']=en.fit_transform(df[["ord_2"]])
df
le = LabelEncoder()
df['Nom_0'] = le.fit_transform(df[["nom_0"]])
df  
from category_encoders import BinaryEncoder
be = BinaryEncoder()
data = be.fit_transform(df['bin_1'])
df = pd.concat([df,data],axis=1)
df
be = BinaryEncoder()
data = be.fit_transform(df['bin_2'])
df = pd.concat([df,data],axis=1)
df

Output:

Initial data

Unique data

Original data

Label Encoder

Binary Encoder

Data.csv

import pandas as pd
import numpy as np
from scipy import stats
import seaborn as sns
import matplotlib.pyplot as plt
data=pd.read_csv("/content/data.csv")
df=pd.DataFrame(data)
df
from sklearn.preprocessing import LabelEncoder, OrdinalEncoder
pm=['Hot','Warm','Cold','Very Hot']
e1=OrdinalEncoder(categories=[pm])
e1.fit_transform(df[["Ord_1"]])
df['bo2']=e1.fit_transform(df[["Ord_1"]])
df
le=LabelEncoder()
dfc=df.copy()
dfc['Ord_1']=le.fit_transform(dfc['Ord_1'])
dfc
from sklearn.preprocessing import OneHotEncoder
ohe=OneHotEncoder(sparse=False)
df2=df.copy()
enc=pd.DataFrame(ohe.fit_transform(df2[['City']]))
df2=pd.concat([df2,enc],axis=1)
pd.get_dummies(df2,columns=["City"])
from category_encoders import BinaryEncoder
be = BinaryEncoder()
data = be.fit_transform(dfc['bin_1'])
dfc = pd.concat([dfc,data],axis=1)
dfc
be1 = BinaryEncoder()
data = be1.fit_transform(dfc['bin_2'])
dfc = pd.concat([dfc,data],axis=1)
dfc