Custom tips

Disable .Rdata saving:
- Tools -> Global Options -> change save .Rdata (never) and disable restore .Rdata

copilot

github settings, activate copilot talonendm
university students, free
Rstudio ohjeet posit.co
- Navigate to Tools > Global Options > Copilot.
- Check the box to “Enable GitHub Copilot”.
- Download and install the Copilot Agent components.
- Click the “Sign In” button.
- In the “GitHub Copilot: Sign in” dialog, copy the Verification Code.

Chat with AI in RStudio

“Interact with Github Copilot and OpenAI’s GPT (ChatGPT) models directly in RStudio. The chattr Shiny add-in makes it easy for you to interact with these and other Large Language Models (LLMs).” rstudio

chattr package R
- check instructions
tillburg.ai chattr

# Run the app
chattr::chattr_app()

and espace with ESC.

Quarto

Like R Markdown, Quarto uses Knitr to execute R code, and is therefore able to render most existing Rmd files without modification.

R tips - ggplot

time series

# ....................................................................
# Rtips_ggplot.R
# ....................................................................
if (!require(dplyr)) install.packages('dplyr')
library(dplyr)

if (!require(ggplot2)) install.packages('ggplot2')
library(ggplot2)
# ....................................................................
# generate random time series:
datetime <- seq(ISOdate(2020, 05, 01), ISOdate(2020, 05, 31), "hours")
df <- data.frame(datetime, values = sample(0:100, length(datetime), rep = TRUE)) # [1]

head(df)

Random data set was generated, see stackoverflow.com.

# line graph: scale_x_datetime  
gg1 <- ggplot(df, aes(x = datetime, y = values)) + geom_line() + 
scale_x_datetime(date_breaks = "5 days", date_labels = "%y/%m/%d") + 
ylab("values [0..100]")

# detect points when values are 0 or 100 [dplyr]
df_max_min <- df %>% dplyr::filter(values == 0 | values == 100)

Annotate points and labels to gg1 object, see stackoverflow.com and stackoverflow.com.

# annotate points and labels: [2 & 3] # https://stackoverflow.com/questions/36541086/adding-an-extra-point-in-a-ggplot2-graph
gg1 <- gg1 + annotate("point", df_max_min$datetime, df_max_min$values, colour="red") + 
  geom_text(data = df_max_min, aes(x= datetime, y= values, label = "extreme")) + 
  ggtitle("time series, points and labels") 


# add horizontal and vertical lines:
gg1 <- gg1 + geom_vline(xintercept = df_max_min$datetime, linetype = "dashed", color = "red", size=1)  
gg1 <- gg1 + geom_hline(yintercept = 50, linetype = "dotted", color = "green", size=1)


# show figure --------
print(gg1 + ggtitle("time series, points and labels"))

Histogram: geom_histogram

Let’s create larger data frame https://plotly.com/.

df <- data.frame(datetime, 
                 values = sample(0:100, length(datetime), rep = TRUE), 
                 cond = factor( rep_len(c("A","B"), length(datetime))), 
                 rating = c(rnorm(50),rnorm(length(datetime) - 50, mean=.8))) # [1]


gg2 <- ggplot(data=df, aes(x=rating, fill=cond)) +
  geom_vline(xintercept=c(-2,2)) +
  geom_histogram(binwidth=.5, alpha = 0.4)


print(gg2) + ggtitle("Histogram")

many ggplots in the same figure

Required packages:

if (!require(here)) install.packages('here')
library(here) # create path for ggsave

if (!require(cowplot)) install.packages('cowplot')
library(cowplot)

if (!require(ggpubr)) install.packages('ggpubr')
library(ggpubr)

if (!require(here)) install.packages('here')
library(here)

First, let’s change theme a bit. Generated picture is used as top image in this blog post.

gg1 <- gg1 + theme(plot.background = element_rect(fill = "#C0C0C0"), 
                   panel.grid.major = element_line(colour = "grey90"), 
                   axis.ticks = element_line(size = 2)) # https://ggplot2.tidyverse.org/reference/theme.html


gg2 <- gg2 + 
  theme(plot.background = element_rect(fill = "#C0C0C3"))  # https://ggplot2.tidyverse.org/reference/theme.html


gg_plots <- ggarrange(gg1, gg2 , 
                          labels = c("A", "B"),
                          ncol = 2, nrow = 1)

gg_plots <- gg_plots + ggtitle("Time series and histogram")

last_plot()$labels$title
paste0(here("/"), last_plot()$labels$title, ".png")

# wide image on blog post ===========
ggsave(filename = paste0(here("/"), "wide_", last_plot()$labels$title, ".png"),
       width = 29, height = 6, units = "cm",  dpi = 300)

# Square image on main page ===========
ggsave(filename = paste0(here("/"), "square_", gg1$labels$title, ".png"), plot = gg1,
       width = 10, height = 10, units = "cm",  dpi = 300)

Tight x-axis with data and time

+ ggplot2::scale_x_datetime(limit=as.Date(limits = c(startdate_time, '2015-07-01')), expand=c(0,0), date_beaks = "1 day" ) + 

stringr

Extracting the last n characters from a string in R
```
str_sub(data, -4, -1) 
```

Experiments

Rjs test - write R inside html

rjs: R in JavaScript quick test in additional html-page: However, I got this message [4.1.2021]:

Mixed Content: The page at 'https://talonendm.github.io/introR' was loaded over HTTPS, but requested an insecure XMLHttpRequest endpoint 'http://kdpsingh.ocpu.io/RendereR/R/renderer'. This request has been blocked; the content must be served over HTTPS.

In addition, _posts are md-files, so I should do something like this: How to write html into md-file?

Shinybulma

This was originally used in safka.me re:teatteri app in mid 2010. See shinybulma github pages: “Bulma: the modern CSS framework that just works.”

Installation, e.g. in ui.R file as

install.packages('devtools')
library(devtools)
devtools::install_github("RinteRface/shinybulma")
library(shinybulma)

References

Vastuuvapauslauseke

Tämä julkaisu on tehty parhaan oman tietämyksemme ja ymmärtämyksemme pohjalta. Olemme tehneet kaiken voitavamme tietojen täsmällisyyden ja oikeellisuuden takaamiseksi, talonendm ei ole vastuussa mistään virheistä tai puutteista tai mistään suorasta, epäsuorasta tai välillisestä vahingosta, joka on aiheutunut tietojen virheellisestä soveltamisesta. Pidätämme oikeuden muutoksiin.

Tags: R package ggplot package ggsave package chattr LLM copilot

R tips - ggplot package

Simple time series and annotate points and labels on it