file_reading_2.py

import pyspark.sql.types as tp

# define the schema
my_schema = tp.StructType([
    tp.StructField(name= 'Batsman',      dataType= tp.IntegerType(),   nullable= True),
    tp.StructField(name= 'Batsman_Name', dataType= tp.StringType(),    nullable= True),
    tp.StructField(name= 'Bowler',       dataType= tp.IntegerType(),   nullable= True),
    tp.StructField(name= 'Bowler_Name',  dataType= tp.StringType(),    nullable= True),
    tp.StructField(name= 'Commentary',   dataType= tp.StringType(),    nullable= True),
    tp.StructField(name= 'Detail',       dataType= tp.StringType(),    nullable= True),
    tp.StructField(name= 'Dismissed',    dataType= tp.IntegerType(),   nullable= True),
    tp.StructField(name= 'Id',           dataType= tp.IntegerType(),   nullable= True),
    tp.StructField(name= 'Isball',       dataType= tp.BooleanType(),   nullable= True),
    tp.StructField(name= 'Isboundary',   dataType= tp.BinaryType(),   nullable= True),
    tp.StructField(name= 'Iswicket',     dataType= tp.BinaryType(),   nullable= True),
    tp.StructField(name= 'Over',         dataType= tp.DoubleType(),    nullable= True),
    tp.StructField(name= 'Runs',         dataType= tp.IntegerType(),   nullable= True),
    tp.StructField(name= 'Timestamp',    dataType= tp.TimestampType(), nullable= True)    
])

# read the data again with the defined schema
my_data = spark.read.csv('ind-ban-comment.csv',schema= my_schema,header= True)

# print the schema
my_data.printSchema()