作者mowgur (PINNNNN)
看板R_Language
标题[问题] 制作dummy variable矩阵效能问题
时间Fri Dec 29 14:01:27 2017
*[m- 问题: 当你想要问问题时，请使用这个类别。
建议先到
http://tinyurl.com/mnerchs 搜寻本板旧文。
[问题类型]:
效能谘询(我想让R 跑更快)
[软体熟悉度]:
使用者(已经有用R 做过不少作品)
[问题叙述]:
大家好 我的资料是纪录篮球比赛每个play是哪5个进攻及防守球员在场上
想做的事情是: 假设总共有500位球员 做出一个n(750000) x p(1000)的矩阵
前500栏为进攻 後500栏为防守
矩阵内的元素为1代表球员在场上进攻(防守为-1) 不在场上为0
所以每列会有5个1及5个-1还有很多个0
资料大概长这样
data$p.combination data$p.com.allowed
1 A, B, C, D, E J, K, L, M, N
2 A, C, F, H, I K, L, M, N, O
3 C, D, X, Y, Z K, M, O, Q, R
... ... ...
人名之间是用逗号和一个空格分开
用我自己写的已经跑了快12小时还没跑完
想请教版上各位大大有没有更好的写法
[程式范例]:
https://ideone.com/PaBtM4
library(magrittr)
p.combination = character(1000)
for(i in 1:length(p.combination)){
p.combination[i] = LETTERS[sample(1:26,5)] %>% paste0(collapse = ", ")
}
p.com.allowed = character(1000)
for(i in 1:length(p.com.allowed)){
p.com.allowed[i] = LETTERS[sample(1:26,5)] %>% paste0(collapse = ", ")
}
data = data.frame(p.combination = p.combination,
p.com.allowed = p.com.allowed)
player = LETTERS[1:26]
input.matrix0 = function(data, player, off){
X = matrix(ncol = length(player), nrow = dim(data)[1])
for(i in 1:dim(data)[1]){
if(off) {
colnames(X) = paste0("O_",player)
coding = 1
pp = data$p.combination
} else {
colnames(X) = paste0("D_",player)
coding = -1
pp = data$p.com.allowed
}
player.temp = pp[i] %>% gsub(", ", "|",.)
index = grep(player.temp, player)
X[i,index] = coding
X[i,-index] = 0
}
return(X)
}
input.matrix = function(data, player){
X.off = input.matrix0(data, player, T)
X.def = input.matrix0(data, player, F)
return(cbind(X.off, X.def))
}
out = input.matrix(data,player)
--
※ 发信站: 批踢踢实业坊(ptt.cc), 来自: 58.114.236.110
※ 文章网址: https://webptt.com/cn.aspx?n=bbs/R_Language/M.1514527291.A.C47.html