2021-03: Binary Diagnostic

library(tidyverse)

── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
✔ ggplot2 3.4.0      ✔ purrr   0.3.5 
✔ tibble  3.1.8      ✔ dplyr   1.0.10
✔ tidyr   1.2.1      ✔ stringr 1.4.1 
✔ readr   2.1.3      ✔ forcats 0.5.2 
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()

library(data.table)

Attaching package: 'data.table'

The following objects are masked from 'package:dplyr':

    between, first, last

The following object is masked from 'package:purrr':

    transpose

library(compositions)

Welcome to compositions, a package for compositional data analysis.
Find an intro with "? compositions"


Attaching package: 'compositions'

The following objects are masked from 'package:stats':

    anova, cor, cov, dist, var

The following objects are masked from 'package:base':

    %*%, norm, scale, scale.default

Part 1

# part 1
dt <- data.frame(fread("input.txt", sep = "\n", colClasses = c("character")))
dt <- tidyr::separate(dt, "V1", paste0("V", 1:13), sep = "", convert = TRUE)

get_mode <- function(x)
  return(names(sort(table(x), decreasing = T, na.last = T)[1]))
o <- c()
for(i in 2:13)
  o <- append(o, get_mode(dt[,i]))

o <- unbinary(paste(o, collapse = ""))

get_mode <- function(x)
  return(names(sort(table(x), na.last = T)[1]))
c <- c()
for(i in 2:13)
  c <- append(c, get_mode(dt[,i]))
c <- unbinary(paste(c, collapse = ""))

o * c

[1] 4174964

Part 2

# part 2
dtM <- data.frame(fread("input.txt", sep = "\n", colClasses = c("character")))
dtM <- tidyr::separate(dtM, "V1", paste0("V", 1:13), sep = "", convert = TRUE)
dtM <- dtM %>% select(-"V1")
dt <- dtM

get_modeMax <- function(x){
  temp <- sort(table(x), decreasing = T, na.last = T)
  ifelse(temp[1] == temp[2], return(1), names(temp[1]))
}
o <- c()

for(i in 1:12) {
  dt <- dt %>%
    filter_at(i, all_vars(.==get_modeMax(dt[,i])))
  if(nrow(dt) == 1) {
    oxygen <- paste(as.character(dt[1,]), collapse = "")
    break
  }
}

dt <- dtM
get_modeMin <- function(x){
  temp <- sort(table(x), na.last = T)
  ifelse(temp[1] == temp[2], return(0), names(temp[1]))
}
c <- c()

for(i in 1:12) {
  dt <- dt %>%
    filter_at(i, all_vars(.==get_modeMin(dt[,i])))
  if(nrow(dt) == 1) {
    co2 <- paste(as.character(dt[1,]), collapse = "")
    break
  }
}

unbinary(oxygen) * unbinary(co2)

[1] 4474944

Speed Edition

Today was bad. I misread the instructions, forgot a bunch of basic functions, and so much more. I couldn’t get any of the base reading functions to read my input properly so I ended up with fread from data.table. I also haven’t been able to make any significant speed improvements from my initial solution.

I’m not sure why readLines wasn’t working last night but I also switched to str_split_fixed from stringr and it’s so much faster now.

Fastest solution

library(dplyr)
# part 1
dt <- stringr::str_split_fixed(readLines("input.txt"), pattern = "", n = 12)
dt <- data.frame(apply(dt, 2, as.integer))

getModeMin <- function(x)
  return(ifelse(mean(x) > .5, 0, 1))
getModeMax <- function(x)
  return(ifelse(mean(x) > .5, 1, 0))

strtoi(paste(apply(dt, 2, getModeMax), collapse = ""), base = 2) *
  strtoi(paste(apply(dt, 2, getModeMin), collapse = ""), base = 2)

[1] 4174964

# part 2
dto <- dt
dtc <- dt

for (i in 1:12) {
  if (nrow(dto) > 1)
    dto <- dto |>
      filter_at(i, all_vars(. == ifelse(mean(dto[, i]) >= .5, 1, 0)))
  if (nrow(dtc) > 1)
    dtc <- dtc |>
      filter_at(i, all_vars(. == ifelse(mean(dtc[, i]) < .5, 1, 0)))
}

strtoi(paste(as.character(dto[1,]), collapse = ""), base = 2) *
  strtoi(paste(as.character(dtc[1,]), collapse = ""), base = 2)

[1] 4474944

Benchmarks

rbenchmark::benchmark(
  "First try" = {
    library(tidyverse)
    library(data.table)
    library(compositions)
    
    # part 1
    dt <- data.frame(fread("input.txt", sep = "\n", 
                           colClasses = c("character")))
    dt <- tidyr::separate(dt, "V1", paste0("V", 1:13), 
                          sep = "", convert = TRUE)
    
    get_mode <- function(x)
      return(names(sort(table(x), decreasing = T, na.last = T)[1]))
    o <- c()
    for (i in 2:13)
      o <- append(o, get_mode(dt[, i]))
    o <- unbinary(paste(o, collapse = ""))
    
    get_mode <- function(x)
      return(names(sort(table(x), na.last = T)[1]))
    c <- c()
    for (i in 2:13)
      c <- append(c, get_mode(dt[, i]))
    c <- unbinary(paste(c, collapse = ""))
    
    o * c
    
    # part 2
    dtM <- data.frame(fread("input.txt", sep = "\n",
                            colClasses = c("character")))
    dtM <- tidyr::separate(dtM, "V1", paste0("V", 1:13),
                           sep = "", convert = TRUE)
    dtM <- dtM %>% select(-"V1")
    dt <- dtM

    get_modeMax <- function(x) {
      temp <- sort(table(x), decreasing = T, na.last = T)
      ifelse(temp[1] == temp[2], return(1), names(temp[1]))
    }
    o <- c()
    for (i in 1:12) {
      dt <- dt %>% filter_at(i, all_vars(. == get_modeMax(dt[, i])))
      if (nrow(dt) == 1) {
        oxygen <- paste(as.character(dt[1, ]), collapse = "")
        break
      }
    }

    dt <- dtM
    get_modeMin <- function(x) {
      temp <- sort(table(x), na.last = T)
      ifelse(temp[1] == temp[2], return(0), names(temp[1]))
    }
    c <- c()
    for (i in 1:12) {
      dt <- dt %>% filter_at(i, all_vars(. == get_modeMin(dt[, i])))
      if (nrow(dt) == 1) {
        co2 <- paste(as.character(dt[1, ]), collapse = "")
        break
      }
    }

    unbinary(oxygen) * unbinary(co2)
    },
    "Second try" = {
      library(tidyverse)
      library(data.table)
      library(compositions)
      
      # part 1
      dt <- fread("input.txt", sep = "\n", colClasses = c("character")) %>%
        as.data.frame() %>%
        tidyr::separate("V1", paste0("V", 1:13), sep = "", convert = TRUE) %>%
        select(-"V1")
      
      getModeMin <- function(x)
        return(names(sort(table(x), na.last = T)[1]))
      getModeMax <- function(x)
        return(names(sort(
          table(x), decreasing = T, na.last = T
        )[1]))
      
      o <- unbinary(paste(apply(dt, 2, getModeMax), collapse = ""))
      
      c <- unbinary(paste(apply(dt, 2, getModeMin), collapse = ""))
      
      o * c
      
      # part 2
      dtM <-
        fread("input.txt", sep = "\n", colClasses = c("character")) %>%
        as.data.frame() %>%
        tidyr::separate("V1", paste0("V", 1:13), sep = "", convert = TRUE) %>%
        select(-"V1")
      dto <- dtM
      dtc <- dtM
      
      getModeMin <- function(x) {
        temp <- sort(table(x), na.last = T)
        ifelse(temp[1] == temp[2], return(0), names(temp[1]))
      }
      getModeMax <- function(x) {
        temp <- sort(table(x), decreasing = T, na.last = T)
        ifelse(temp[1] == temp[2], return(1), names(temp[1]))
      }
      
      for (i in 1:12) {
        if (nrow(dto) > 1)
          dto <-
            dto %>% filter_at(i, all_vars(. == getModeMax(dto[, i])))
        if (nrow(dtc) > 1)
          dtc <-
            dtc %>% filter_at(i, all_vars(. == getModeMin(dtc[, i])))
        if (nrow(dto) == 1 & nrow(dtc) == 1) {
          oxygen <- paste(as.character(dto[1,]), collapse = "")
          co2 <- paste(as.character(dtc[1,]), collapse = "")
          break
        }
      }
      
      unbinary(oxygen) * unbinary(co2)
    }, 
  "Third try" = {
    library(dplyr)
    # part 1
    dt <- 
      stringr::str_split_fixed(readLines("input.txt"), pattern = "", n = 12)
    dt <- data.frame(apply(dt, 2, as.integer))
    
    getModeMin <- function(x)
      return(ifelse(mean(x) > .5, 0, 1))
    getModeMax <- function(x)
      return(ifelse(mean(x) > .5, 1, 0))
    
    strtoi(paste(apply(dt, 2, getModeMax), collapse = ""), base = 2) *
      strtoi(paste(apply(dt, 2, getModeMin), collapse = ""), base = 2)
    
    # part 2
    dto <- dt
    dtc <- dt
    
    for (i in 1:12) {
      if (nrow(dto) > 1)
        dto <- dto |>
          filter_at(i, all_vars(. == ifelse(mean(dto[, i]) >= .5, 1, 0)))
      if (nrow(dtc) > 1)
        dtc <- dtc |>
          filter_at(i, all_vars(. == ifelse(mean(dtc[, i]) < .5, 1, 0)))
    }
    
    strtoi(paste(as.character(dto[1, ]), collapse = ""), base = 2) *
      strtoi(paste(as.character(dtc[1, ]), collapse = ""), base = 2)
  }, 
  replications = 100, columns = c(1:5), order = "user.self")

        test replications user.self sys.self elapsed
2 Second try          100    16.956    0.346  17.304
1  First try          100    17.130    0.360  17.492
3  Third try          100    20.403    0.318  20.740