Explore the Advantages of R, Spark, and sparklyrR is mostly optimized to help you write data analysis code quickly and readably. Apache Spark is designed to analyze huge datasets quickly. The sparklyr package lets you write dplyr R code that runs on a Spark cluster, giving you the best of both worlds. This 4-hour course teaches you how to manipulate Spark DataFrames using both the dplyr interface and the native interface to Spark, as well as trying machine learning techniques.
Load Data into Spark and Manipulate Spark DataFramesYou’ll start this Spark course by investigating how Spark and R work well together and practicing loading data, ready for cleaning, transformation, and analysis. You’ll use Spark frames and dplyr syntax to manipulate your data by filtering and arranging rows, and mutating and summarizing columns.
Delve into Big Data Analysis with Spark MLibThis course focuses on building your skills and confidence in analyzing huge datasets. The final chapters take you through Spark’s machine learning data transformation features and offer you the chance to practice sparklyr’s machine learning routines by using it to make predictions using gradient boosted trees and random forests. "
Light My Fire: Starting To Use Spark With dplyr SyntaxFree
In which you learn how Spark and R complement each other, how to get data to and from Spark, and how to manipulate Spark data frames using dplyr syntax.Getting started50 xpMade for each other50 xpHere be dragons50 xpThe connect-work-disconnect pattern100 xpCopying data into Spark100 xpBig data, tiny tibble100 xpExploring the structure of tibbles100 xpSelecting columns100 xpFiltering rows100 xpArranging rows100 xpMutating columns100 xpSummarizing columns100 xp
Tools of the Trade: Advanced dplyr Usage
In which you learn more about using the
dplyrinterface to Spark, including advanced field selection, calculating groupwise statistics, and joining data frames.Leveling up50 xpMother's little helper (1)100 xpMother's little helper (2)100 xpSelecting unique rows100 xpCommon people100 xpCollecting data back from Spark100 xpStoring intermediate results100 xpGroups: great for music, great for data100 xpGroups of mutants100 xpAdvanced Selection II: The SQL100 xpLeft joins100 xpAnti joins100 xpSemi joins100 xp
Going Native: Use The Native Interface to Manipulate Spark DataFrames
In which you learn about Spark's machine learning data transformation features, and functionality for manipulating native DataFrames.Two new interfaces50 xpPopcorn double feature50 xpTransforming continuous variables to logical100 xpTransforming continuous variables into categorical (1)100 xpTransforming continuous variables into categorical (2)100 xpMore than words: tokenization (1)100 xpMore than words: tokenization (2)100 xpMore than words: tokenization (3)100 xpSorting vs. arranging100 xpExploring Spark data types100 xpShrinking the data by sampling100 xpTraining/testing partitions100 xp
Case Study: Learning to be a Machine: Running Machine Learning Models on Spark
A case study in which you learn to use
sparklyr's machine learning routines, by predicting the year in which a song was released.Machine learning on Spark50 xpMachine learning functions50 xp(Hey you) What's that sound?100 xpWorking with parquet files100 xpCome together100 xpPartitioning data with a group effect100 xpGradient boosted trees: modeling100 xpGradient boosted trees: prediction100 xpGradient boosted trees: visualization100 xpRandom Forest: modeling100 xpRandom Forest: prediction100 xpRandom Forest: visualization100 xpComparing model performance100 xp
DatasetsAnti-joinBoth-model-responsesGbt-model-responsesInner-joinLeft-joinPredicted vs actualResidual densitySemi-joinTimbreTimbre parquetTitle text parquetTrack data parquetTrack data to model parquetTrack data to predict parquetTrack metadata
Richie CottonSee More
Data Evangelist at DataCamp
Richie is a Data Evangelist at DataCamp. He has been using R since 2004, in the fields of proteomics, debt collection, and chemical health and safety. He has released almost 30 R packages on CRAN and Bioconductor – most famously the assertive suite of packages – as well as creating and contributing to many others. He also has written two books on R programming, Learning R and Testing R Code.