Laurae2

Code:

library(anytime)
library(flipTime)
library(data.table)

set.seed(1)
system.time({
  some_dates <- as.Date("2020-12-31") - sample(40000, 1e7, TRUE)

Laurae2

Walk on thin ice in R for gcc-7/g++-7 + CUDA 10.0.130

Note: OpenCL can run on Single Precision (gpu_use_dp = FALSE) or Double Precision (gpu_use_dp = TRUE) whereas CUDA is strictly Double Precision in LightGBM.

--

STEP 1: HACK IN SOME FILES

Hack in the following:
```

Laurae2

library(xgboost)
library(lightgbm)
library(data.table)

setwd("/home/laurae/Documents/R/GBM-perf")

n_threads <- 16

data <- fread("HIGGS.csv")
labels <- data$V1

Laurae2

library(xgboost)

set.seed(1)
N <- 1000000
p <- 100
pp <- 25
X <- matrix(runif(N * p), ncol = p)
betas <- 2 * runif(pp) - 1
sel <- sort(sample(p, pp))
m <- X[, sel] %*% betas - 1 + rnorm(N)

Laurae2

set.seed(1)
N <- 500000
p <- 100
pp <- 25
X <- matrix(runif(N * p), ncol = p)
betas <- 2 * runif(pp) - 1
sel <- sort(sample(p, pp))
m <- X[, sel] %*% betas - 1 + rnorm(N)
y <- rbinom(N, 1, plogis(m))

Laurae2

library(xgboost)

set.seed(1)
N <- 10000000
p <- 100
pp <- 25
X <- matrix(runif(N * p), ncol = p)
betas <- 2 * runif(pp) - 1
sel <- sort(sample(p, pp))
m <- X[, sel] %*% betas - 1 + rnorm(N)

Laurae2

Step 1: download cplexAPI : Step 2: edit src/Makevars.win like the following, make sure cplex1280 is equal to your version you must change to the corresponding number you can find in C:/Program Files/IBM/ILOG/CPLEX_Studio128/cplex/bin/x64_win64:

PKG_CPPFLAGS=-g -D_R_=1 -DUSE_R=1 -I"C:/Program Files/IBM/ILOG/CPLEX_Studio128/cplex/include"
PKG_LIBS=-L"C:/Program Files/IBM/ILOG/CPLEX_Studio128/cplex/bin/x64_win64" -lm -lcplex1280

Step 3: in a R console, on the right folder:

Laurae2

library(Rcpp)

Sys.setenv(PKG_CXXFLAGS = "-fopenmp -std=c++11")

cppFunction(code = '#include <Rcpp.h>
#include <omp.h>

Rcpp::NumericVector streamer(Rcpp::NumericVector threads, int triad_size = 2147483647, int repeats = 10, int repeats_alloc = 3, bool simd = false) {
  
  int i;

Laurae2

# library(data.table)
# N=2e9; K=100
# set.seed(1)
# DT <- data.table(
#   id1 = sample(sprintf("id%03d",1:K), N, TRUE),      # large groups (char)
#   id2 = sample(sprintf("id%03d",1:K), N, TRUE),      # large groups (char)
#   id3 = sample(sprintf("id%010d",1:(N/K)), N, TRUE), # small groups (char)
#   id4 = sample(K, N, TRUE),                          # large groups (int)
#   id5 = sample(K, N, TRUE),                          # large groups (int)
#   id6 = sample(N/K, N, TRUE),                        # small groups (int)

Laurae2

# List all possible power config GUIDs in Windows
# Run: this-script.ps1 | Out-File powercfg.ps1
# Then edit and run powercfg.ps1
# (c) Pekka "raspi" Järvinen 2017

$powerSettingTable = Get-WmiObject -Namespace root\cimv2\power -Class Win32_PowerSetting
$powerSettingInSubgroubTable = Get-WmiObject -Namespace root\cimv2\power -Class Win32_PowerSettingInSubgroup

Get-WmiObject -Namespace root\cimv2\power -Class Win32_PowerSettingCapabilities | ForEach-Object {
  $tmp = $_.ManagedElement