Zeppelin and SparkR

#Install R and then following packages
#repr failed to create 
yum install R-*
install.packages("evaluate", dependencies = TRUE)
install.packages("base64enc", dependencies = TRUE)
install.packages("devtools", dependencies = TRUE)
install_github('IRkernel/repr')
install.packages("dplyr", dependencies = TRUE)
install.packages("caret", dependencies = TRUE)
install.packages("repr", dependencies = TRUE)

#setup zeppelin
git clone https://github.com/elbamos/Zeppelin-With-R
mvn clean package -DskipTests

#Notebook 
%spark.r
library(magrittr)
library(dplyr)
library(ggplot2)
library(caret)
# flights.csv is on local os 
%spark.r
flights <- read.csv("/tmp/flights.csv") 
head(flights)
## dataset https://github.com/SparkIQ-Labs/Demos/tree/master/datasets/nyc-flights-dataset

%spark.r
# mean arrival delays for each destination airport
flights %>%
group_by(dest) %>%
summarise(
    arr_delay = mean(arr_delay, na.rm = TRUE),
    n=n()) %>%
    arrange(desc(arr_delay))
# plot
%spark.knitr

``` {r echo = FALSE}
per_hour <- flights %>%
filter (cancelled == 0) %>%
mutate(time = hour + minute / 60) %>%
group_by(time) %>%
summarise(
    arr_delay = mean(arr_delay, na.rm = TRUE),
    n = n()
)

ggplot(filter(per_hour, n > 30 ), aes(time, arr_delay)) +
geom_vline(xintercept = 5:24 , colour = "white", size = 2) + 
geom_point()
```