Project

General

Profile

« Previous | Next » 

Revision ec229367

Added by Benoit Parmentier over 12 years ago

  • ID ec229367f7438821b4823ffad7fd351ac6d89f5b

Streamlining and slight changes to the loop to create subset dataset for different dates

View differences:

climate/research/oregon/interpolation/Linear_reg.R
22 22 

  		




  23
  23
  
    
###Reading the station data


  		




  24
  24
  
    
ghcn<-read.csv(paste(path,"/",infile1, sep=""), header=TRUE)                            #The "paste" function concatenates the path and file name in common string. 


  		




  25
  
  
    
#ghcn<-read.csv(infile1)                                                                #Use read.table if the input file is space delimited or read.table(infile1, headername=TRUE, sep=',')                                                                            #Checking that the columns are correctly labelled.


  		




  26
  25
  
    
dates <-readLines(paste(path,"/",infile2, sep=""))


  		




  27
  26
  
    
                  


  		




  28
  27
  
    
###Creating a validation dataset by creating training and testing datasets (%30)


  		




  29
  
  
    
#ghcn1507 <-subset(ghcn,ghcn$date_== date)


  		




  30
  
  
    
#for loops


  		




  31
  
  
    
ddat <- as.list(rep("", length(dates)))


  		




  
  28
  
    

  		




  
  29
  
    

  		




  
  30
  
    
ghcn.subsets <-lapply(dates, function(d) subset(ghcn, date_==d))


  		




  32
  31
  
    

  		




  33
  32
  
    
for(i in 1:length(dates)){


  		




  34
  
  
    
  #assign(paste("ddat",i,sep="_"),data)


  		




  35
  
  
    
  data_name<-cat("ghcn_",dates[[1]],sep="")


  		




  36
  
  
    
  #data<-as.data.frame(data)


  		




  37
  
  
    
  #ddat[[i]] <- data.frame(subset(ghcn,ghcn$date_==dates[[i]]))


  		




  38
  
  
    
  #data<-subset(ghcn,ghcn$date_==dates[[i]])


  		




  39
  
  
    
  data<-subset(ghcn,ghcn$date_==is.numeric(dates[[i]])


  		




  
  33
  
    
  #data_name<-cat("ghcn_",dates[[i]],sep="")


  		




  
  34
  
    
  data_name<-paste("ghcn_",dates[[i]],sep="")


  		




  
  35
  
    
  data<-subset(ghcn,ghcn$date_==dates[[i]])


  		




  40
  36
  
    
  assign(data_name,data)


  		




  41
  
  
    
  n<-nrow(data[[i]])


  		




  42
  
  
    
  ns<-n-round(n*0.3)  #Create a sample from the data frame with 70% of the rows


  		




  43
  
  
    
  #ns<-n-round(n*prop)  #Create a sample from the data frame with 70% of the rows


  		




  44
  
  
    
  ind.training <- sample(nrow(data), size=ns, replace=FALSE) #This selects the index position for 70% of the rows taken randomly


  		




  45
  
  
    
  ind.testing <- setdiff(1:nrow(data), ind.training)


  		




  46
  
  
    
  data_s <- data[ind.training, ]


  		




  47
  
  
    
  data_v <- data[ind.testing, ]


  		




  48
  
  
    
  }


  		




  
  37
  
    
  }                                                   #This loops creates 2 subsets of ghcn based on dates and reassign the names using the date


  		




  49
  38
  
    

  		




  50
  39
  
    
# ############ REGRESSION ###############


  		




  51
  40
  
    
# 


  		












Also available in:  Unified diff