val ng- pan dsm reg output turn 輸出 out

library(rhdfs) library(rmr2) hdfs.init() hdfs.delete("/user/output/lm.output") map <- function(k,lines) { lines<-unlist(strsplit(lines,‘#‘)) k<-lines[1] x<-unlist(strsplit(lines[2],‘,‘)) y<-unlist(strsplit(lines[3],‘,‘)) x<-as.numeric(x) y<-as.numeric(y) lm <- lm(y ~ x) return( keyval(k, lm$coefficients[[2]]) ) } reduce <- function(key, lmres) { # keyval(key, lmres) return lmres } wordcount <- function (input, output=NULL) { mapreduce(input=input, output=output, input.format="text",output.format = "text", map=map, reduce=reduce) } hdfs.root <- ‘/user‘ hdfs.data <- file.path(hdfs.root, ‘input/lm.input‘) hdfs.out <- file.path(hdfs.root, ‘output/lm.output‘) out <- wordcount(hdfs.data, hdfs.out)
輸入文件為：/user/input/lm.input 1#1,2,3,4#2,4,6,8 2#1,2,3,4#3,6,9,12 3#1,2,3,4#4,8,12,16 4#1,2,3,4#1.5,3,4.5,6
可是輸出卻僅僅有2個結果： 1 2 4 1.5

rhadoop linear regression 問題