Skip to content

Latest commit

 

History

History
21 lines (18 loc) · 1.02 KB

README.md

File metadata and controls

21 lines (18 loc) · 1.02 KB

CompareMLLibWithR

In this project, we compared two implementations of the Lasso (L1) regression algorithm:

  1. the R function cv.glmnet() in the glmnet package,
  2. the function LassoWithSGD() in MLlib.

Our objective was to compare the two algorithms in terms of:

  1. usability,
  2. performance,
  3. accuracy.

The following files are part of this project:

File Content
CC.R The main R script that both creates the input files and analyses each file with the cv.glmnet() function.
properties.R A properties file that controls the latin hypercube sampling.
workingDir.R This file is not submitted because it contains the hardcoded path variables of the local user.
ScalaShell.scala A scala script file that can be run in the Spark shell.
CompareResultsLasso_CompareSparkvsR.csv A csv file with the combined results of the R and Spark simulations.
analyzeResults.R A script to analyze the above scv file.
Spark_versus_R.pdf A poster that summerizes the results of the project.