Skip to content
Related Articles

Related Articles

Pyspark | Linear regression using Apache MLlib

View Discussion
Improve Article
Save Article
  • Difficulty Level : Medium
  • Last Updated : 19 Jul, 2019

Problem Statement: Build a predictive Model for the shipping company, to find an estimate of how many Crew members a ship requires.
The dataset contains 159 instances with 9 features.

The Description of dataset is as below:

Let’s make the Linear Regression Model, predicting Crew members

Attached dataset: cruise_ship_info

import pyspark
from pyspark.sql import SparkSession
#SparkSession is now the entry point of Spark
#SparkSession can also be construed as gateway to spark libraries
#create instance of spark class
#create spark dataframe of input csv file'D:\python coding\pyspark_tutorial\Linear regression\cruise_ship_info.csv'

Output :

|  Ship_name|Cruise_line|Age|           Tonnage|passengers|length|cabins|passenger_density|crew|
|    Journey|    Azamara|  6|30.276999999999997|      6.94|  5.94|  3.55|            42.64|3.55|
|      Quest|    Azamara|  6|30.276999999999997|      6.94|  5.94|  3.55|            42.64|3.55|
|Celebration|   Carnival| 26|            47.262|     14.86|  7.22|  7.43|             31.8| 6.7|
|   Conquest|   Carnival| 11|             110.0|     29.74|  9.53| 14.88|            36.99|19.1|
|    Destiny|   Carnival| 17|           101.353|     26.42|  8.92| 13.21|            38.36|10.0|
|    Ecstasy|   Carnival| 22|            70.367|     20.52|  8.55|  10.2|            34.29| 9.2|
|    Elation|   Carnival| 15|            70.367|     20.52|  8.55|  10.2|            34.29| 9.2|
|    Fantasy|   Carnival| 23|            70.367|     20.56|  8.55| 10.22|            34.23| 9.2|
|Fascination|   Carnival| 19|            70.367|     20.52|  8.55|  10.2|            34.29| 9.2|
|    Freedom|   Carnival|  6|110.23899999999999|      37.0|  9.51| 14.87|            29.79|11.5|

#prints structure of dataframe along with datatype

Output :

#In our predictive model, below are the columns

Output :

#columns identified as features are as below:
#to work on the features, spark MLlib expects every value to be in numeric form
#feature 'Cruise_line is string datatype
#using StringIndexer, string type will be typecast to numeric datatype
#import library strinindexer for typecasting
from import StringIndexer
#above code will convert string to numeric feature and create a new dataframe
#new dataframe contains a new feature 'cruise_cat' and can be used further
#feature cruise_cat is now vectorized and can be used to fed to model
for item in indexed.head(5):

Output :

Row(Ship_name='Journey', Cruise_line='Azamara', Age=6, 
Tonnage=30.276999999999997, passengers=6.94, length=5.94, 
cabins=3.55, passenger_density=42.64, crew=3.55, cruise_cat=16.0)

Row(Ship_name='Quest', Cruise_line='Azamara', Age=6, 
Tonnage=30.276999999999997, passengers=6.94, length=5.94, 
cabins=3.55, passenger_density=42.64, crew=3.55, cruise_cat=16.0)

Row(Ship_name='Celebration', Cruise_line='Carnival', Age=26, 
Tonnage=47.262, passengers=14.86, length=7.22, 
cabins=7.43, passenger_density=31.8, crew=6.7, cruise_cat=1.0)

Row(Ship_name='Conquest', Cruise_line='Carnival', Age=11, 
Tonnage=110.0, passengers=29.74, length=9.53,
 cabins=14.88, passenger_density=36.99, crew=19.1, cruise_cat=1.0)

Row(Ship_name='Destiny', Cruise_line='Carnival', Age=17, 
Tonnage=101.353, passengers=26.42, length=8.92, 
cabins=13.21, passenger_density=38.36, crew=10.0, cruise_cat=1.0)

from import Vectors
from import VectorAssembler
#creating vectors from features
#Apache MLlib takes input if vector form
#output as below

Output :

#final data consist of features and label which is crew.'features','crew')
#splitting data into train and test

Output :


Output :

#import LinearRegression library
from import LinearRegression
#creating an object of class LinearRegression
#object takes features and label as input arguments
#pass train_data to train model
#evaluating model trained for Rsquared error
print('Rsquared Error :',ship_results.r2)
#R2 value shows accuracy of model is 92%
#model accuracy is very good and can be use for predictive analysis

Output :

#testing Model on unlabeled data
#create unlabeled data from test_data
#testing model on unlabeled data'features')

Output :

#below are the results of output from test data

Output :

My Personal Notes arrow_drop_up
Recommended Articles
Page :

Start Your Coding Journey Now!