logTrans.Rmd

---
title: "Log Transforming the Data"
author: "Xingyao Chen"
date: "6/21/2017"
output:
  pdf_document: default
  html_document: default
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

```{r}

mydat_small=read.csv('pollinator_visitation_fullData.csv')


mydat_small$Experiment.Week=as.factor(mydat_small$Experiment.Week)
mydat_small$Plant.Number=as.factor(mydat_small$Plant.Number)
mydat_small$Pair=as.factor(mydat_small$Pair)



```
#Make some adjustments before Log10-ing
```{r}
mydat_small=mydat_small[,-1]
boo=c()
for (i in 1:ncol(mydat_small)){
  boo[i]=is.numeric(mydat_small[,i])
}

#Cannot take the log of 0, so added a small value to each column that contains 0's
mydat_small_val_log=mydat_small
mydat_small_val_log$Honeybees=mydat_small_val_log$Honeybees+1
mydat_small_val_log$Speed=mydat_small_val_log$Speed+0.1

#log10 transformation happenning
for(i in 1:ncol(mydat_small_val_log)){
  if(boo[i])
    mydat_small_val_log[,i]=log10(mydat_small_val_log[,i])
}
#peek
summary(mydat_small_val_log)


data_dt=split(mydat_small_val_log, mydat_small$Date)
###median as theshold value
theshold=c()
wholedf=data.frame()
for( i in 1:length(data_dt)){
  theshold[i]=median(data_dt[[i]]$Honeybees)
  data_dt[[i]]$visits=(data_dt[[i]]$Honeybees>=theshold[i])
  wholedf=rbind(wholedf, data_dt[[i]])
}
mydat_small_val_log$visits=wholedf$visits
head(mydat_small_val_log)
write.csv(mydat_small_val_log,'pollinator_visitation_fullData_logTrans.csv')
```