Ggplot - Histogram (geom_histogram, geom

About

Data Visualisation - Histogram (Frequency distribution)

Articles Related

Example

Numeric

library(ggplot2)
library(plotly)
histoTotalTime = ggplot(res_succes, aes(res_succes$TOTAL_TIME_SEC)) + 
  geom_histogram(bins = 100) + 
  labs(x = "Query Time (Sec)", y = "Number of Query") 

ggplotly(histoTotalTime)

Time

R - Date Time (POSIXct)

end_ts_time=as.POSIXct(res_succes$END_TS, format="%Y-%m-%d %H:%M:%S", tz="UTC")
ggplotly(ggplot(res_succes, aes(x=end_ts_time, color=factor(res_succes$USER_NAME)) ) 
         + geom_freqpoly(binwidth = 30)
         + scale_x_datetime()
         + labs(title="Request Frequency Diagram", color="Threads", x="Request End Time", y="Request Count"))

Formula

current/geom_histogram.html = stat_bin + geom_bar + position = “stacked”
geom_freqpoly = stat_bin + geom_line + position = “identity”

Breaks definition

The below code handles outliers by:

creating manually the breaks
limiting the Cartesian coordinates (zooming)

## Create bin breaks
value_breaks = c( seq(10,120,by=10), max(res_succes$TOTAL_TIME_SEC))
## Labels
label_breaks = c(as.character(seq(10, 120, by=10)), "Max+")

ggplot(res_succes, aes(x=res_succes$TOTAL_TIME_SEC, fill = factor(res_succes$REPORT_TYPE))) +
  geom_histogram(breaks=value_breaks) +
  labs(x = "Total Time (min)", fill= "Report Type") +
  coord_cartesian(xlim=c(10,130)) +
  scale_x_continuous(breaks=value_breaks, labels=label_breaks)