jtilly

library(xgboost)
library(dplyr)

params = list(min_child_weight = 0.00001, lambda = 0 )
nrounds = 1

# sparse ---

jtilly

library(lightgbm)
library(tidyverse)

rm(list = ls())

# We load the default iris dataset shipped with R
data(iris)

iris = as_data_frame(iris) %>% 
    mutate(Species = as.numeric(factor(Species)) - 1) %>% 

jtilly

library(xgboost)
set.seed(1234)

N = 1000
x1 <- runif(N)
x <- ifelse(x1 <= 0.2, as.numeric(NA), x1)
y <- as.numeric(x1 >= 0.9)

bst <- xgboost(data = matrix(x, ncol=1), label = y,
               objective = "binary:logistic", eval_metric = "logloss",

jtilly

rm(list = ls())

dict.orig = unique(readLines("https://raw.githubusercontent.com/first20hours/google-10000-english/master/google-10000-english.txt"))

txt2numeric = function(word.orig, return.orig = TRUE) {
    word = tolower(word.orig)
    word = gsub("([^a-z]){1}", 1, word)
    word = gsub("(a|b|c){1}", 2, word)
    word = gsub("(d|e|f){1}", 3, word)
    word = gsub("(g|h|i){1}", 4, word)

jtilly

#ifndef print_mat_h
#define print_mat_h

void print_mat(arma::mat my_matrix) {
    
    uint cols = my_matrix.n_cols;
    uint rows = my_matrix.n_rows;
    
    Rcout << "--------\n";
    for(uint rX = 0; rX < rows; rX++) {

jtilly

// timer.h
#include <iostream>
#include <sstream>

#ifndef timer_h
#define timer_h

class timer {
    
    private:

jtilly

# making mixture models numerically robust
set.seed(4)
rm(list = ls())

nobs = 1000
alpha = 0.1
p = runif(nobs, min = 0.0, max = 1.0)
q = runif(nobs, min = 0.0, max = 1.0)

# naive computation of log-likelihood contribution

jtilly

#!/bin/bash

 convert -density 300 $1 /tmp/page_%04d.jpg
 convert /tmp/page_*.jpg $2
 rm /tmp/page_*.jpg
 

jtilly

#!/bin/bash
latexmk -pdf -pvc $1

jtilly

alias time='/usr/bin/time -f "\nCPU: %Us\tReal: %es\tRAM: %MKB"'