Skip to content

Instantly share code, notes, and snippets.

View arunsrinivasan's full-sized avatar

Arun Srinivasan arunsrinivasan

215 followers · 0 following
View GitHub Profile
@arunsrinivasan
arunsrinivasan / dplyr_vs_data.table_1.8.11_less_groupings.R
Created December 7, 2013 17:44
Benchmarking dplyr and data.table 1.8.11 commit 1048 (with lesser groups)
# version 1.8.11 (commit 1048)
require(data.table)
# Loading required package: data.table
# data.table 1.8.11 For help type: help("data.table")
## create a huge data.table:
## -------------------------
set.seed(1)
N <- 2e7 # size of DT
@arunsrinivasan
arunsrinivasan / CologneR.R
Last active December 30, 2015 14:09
Script used to generate results for CologneR user group meet (for reproducibility)
require(reshape2)
# data.table commit (1048)
require(data.table)
# Loading required package: data.table
# data.table 1.8.11 For help type: help("data.table")
set.seed(1)
N <- 2e7 # size of DT
@arunsrinivasan
arunsrinivasan / dplyr_vs_data.table_1.8.11.R
Last active December 30, 2015 13:59
Benchmarking dplyr and data.table 1.8.11 commit 1048
# version 1.8.11 (commit 1048)
require(data.table)
# Loading required package: data.table
# data.table 1.8.11 For help type: help("data.table")
## create a huge data.table:
## -------------------------
set.seed(1)
N <- 2e7 # size of DT
@arunsrinivasan
arunsrinivasan / DT_1.8.11_1048_benchmark.R
Last active December 30, 2015 13:38
1.8.11 (commit 1048) : Benchmark: comparison between data.table 1.8.10 and 1.8.11 commit 1048
# version 1.8.11 (commit 1048)
require(data.table)
# Loading required package: data.table
# data.table 1.8.11 For help type: help("data.table")
## create a huge data.table:
## -------------------------
set.seed(1)
N <- 2e7 # size of DT
@arunsrinivasan
arunsrinivasan / DT_1.8.10_benchmark.R
Last active December 30, 2015 12:59
1.8.10 : Benchmark: comparison between data.table 1.8.10 and 1.8.11 commit 1048
# version 1.8.10
require(data.table)
# Loading required package: data.table
# data.table 1.8.10 For help type: help("data.table")
## create a huge data.table:
## -------------------------
set.seed(1)
N <- 2e7 # size of DT
@arunsrinivasan
arunsrinivasan / rbind_fill_benchmarking
Last active October 17, 2016 22:14
data.table version of rbind.fill benchmarking with plyr version of rbind.fill
# The post with benchmarking results is the link given below:
# http://stackoverflow.com/questions/18003717/is-there-any-efficient-way-than-rbind-filllist/18004698#18004698
# This is the script with which the benchmarking and plots were generated in case anyone else wants to replicate it.
# Note: it takes about 2-3 hours for the benchmarking to finish.
require(plyr)
require(data.table)
require(ggplot2)
require(microbenchmark)