So far, we have been using Spark SQL and DataFrames through the Spark shell. To use it in standalone programs, you will need to create it explicitly, from a Spark context:

val conf = new SparkConf().setAppName("applicationName")
val sc = new SparkContext(conf)
val sqlContext = new org.apache.spark.sql.SQLContext(sc)

Additionally, importing the implicits object nested in sqlContext allows the conversions of RDDs to DataFrames:

import sqlContext.implicits._

We will use DataFrames extensively in the next chapter to manipulate data to get it ready for use with MLlib.