fmvcd-plot-nb-v0.Rmd

---
title: "Plot - filling in missing values in climate data"
output:
  html_document: default
  html_notebook: default
---

```{r read_cli}
rm(list=ls())
setwd("C:/Users/teipolykom/git/fmvcd/krest/"); getwd()
#setwd("Z:/home/larisa/Dropbox/git/fmvcd/krest/"); getwd()
#setwd("/home/larisa/Dropbox/git/fmvcd/krest/"); getwd()
file_list <- list.files(pattern = "\\.cli$"); print(file_list)

 for (file in file_list){
  # if the merged dataset does exist, append to it
  if (exists("dataset")){
    temp_dataset <-read.table(file, header=FALSE, sep="")
    dataset<-rbind(dataset, temp_dataset)
    rm(temp_dataset)
  }
  # if the merged dataset doesn't exist, create it
  if (!exists("dataset")){
    dataset <- read.table(file, header=FALSE, sep="")
  }
}
names(dataset) <- c("day", "month", "year", "prec", "temp")
dataset_krest <- dataset
```
####
```{r plot_all_cli}
dataset[dataset$prec == -9999, 4] <- NA; 
dataset[dataset$temp == -9999, 5] <- NA; summary(dataset)
temp_plot <- data.frame(Days=(1:dim(dataset)[1]), Temp=dataset[, 5]/10.)
plot(temp_plot, col="red")
prec_plot <- data.frame(Days=(1:dim(dataset)[1]), Temp=dataset[, 4]/10.)
plot(prec_plot, col="blue")
```
####
```{r plot_year}
# Extracing only Unique Rows based on only 3 Column
matches <- unique(dataset[, 3], incomparables = FALSE, fromLast = FALSE)
for (i in matches){
  plot(1: length(dataset[dataset$year == i, 4]), dataset[dataset$year == i, 4]/10., col="blue", main=paste0("Year: ", i),  xlab="Days", ylab="Prec") 
plot(1: length(dataset[dataset$year == i, 4]), dataset[dataset$year == i, 4]/10., col="blue", main=paste0("Year: ", i),  xlab="Days", ylab="Prec", ylim=c(0, 15))  

print(summary(dataset[dataset$year == i, 4]/10.))
  plot(1: length(dataset[dataset$year == i, 4]), dataset[dataset$year == i, 5]/10., col="red", main=paste0("Year: ", i),  xlab="Days", ylab="Temp")
print(summary(dataset[dataset$year == i, 5]/10.))
}
```
####
```{r link}
# How do you turn the output of a nnet neural network model into an equation?
# http://stats.stackexchange.com/questions/163553/how-do-you-turn-the-output-of-a-nnet-neural-network-model-into-an-equation
```