Second Y-Axis in a R plotly graph - r

I combined 2 charts and I am trying to add the second y-axis, but every time I add the yaxis = "y2" to my code, I lose have the my bar graphs.
> MediaDate Spend Search_Visits Other_Visits MediaDate2
> 2016-04-01 $39654.36 19970 2899 Apr 2016
> 2016-05-01 $34446.28 14460 2658 May 2016
> 2016-06-01 $27402.36 12419 2608 Jun 2016
my original code is:
p <- plot_ly(x= w$MediaDate2,y=w$Search_Visits,name = "Paid Search",
type = "bar")
p2 <- add_trace(p, x=w$MediaDate2, y=w$Other_Visits,name = "Other Traffic",
type = "bar")
spend_visits <- layout(p2, barmode = "stack")
spendvisits2 <- spend_visits %>% add_trace(data=w, x=MediaDate2, y=round(Spend,0), fill="tonexty", mode="lines",
text=w$MediaDate2, hoverinfo='name+y+text', name="Spend")
When I add the yaxis= "y2", only the area chart remains:
`spendvisits2 <- spend_visits %>% add_trace(data=w, x=MediaDate2, y=round(Spend,0), yxis="y2" fill="tonexty", mode="lines",
text=w$MediaDate2, hoverinfo='name+y+text', name="Spend")`
Any suggestions would be immensely helpful. Thank you


See this quick Plotly tutorial for multiple axes. You need to specify the attributes for the second y axis in layout().
df <- data.frame(MediaDate = as.Date(c("2016-04-01","2016-05-01","2016-06-01"), format = "%Y-%m-%d"),
Spend = c(39654, 34446, 27402),
Visits = c(19970, 14450, 12419))
plot_ly(df, x = ~MediaDate, y = ~Spend, type = "bar", name = "Spend") %>%
add_trace(x = ~MediaDate, y = ~Visits, mode = "lines", yaxis = "y2", name = "Visits") %>%
layout(yaxis2 = list(overlaying = "y", side = "right"))


Try this
plot_ly(df) %>%
add_trace(x = ~MediaDate, y = ~Spend,type = "bar", name = "Spend") %>%
add_trace(x = ~MediaDate, y = ~Visits, mode = "lines", yaxis = "y2", name = "Visits") %>%
layout(yaxis2 = list(overlaying = "y", side = "right"))


Adding type='scatter' to what Vance Lopez commented worked for me:
plot_ly(df, x = ~MediaDate, y = ~Spend, type = "bar", name = "Spend") %>%
add_trace(x = ~MediaDate, y = ~Visits, type = 'scatter', mode = "lines", yaxis = "y2", name = "Visits") %>%
layout(yaxis2 = list(overlaying = "y", side = "right"))

Related

The R plot_ly stacked bar chart can only show multiple values with different colors

I am programming a R-application, where I need to visualize data in stacked bar-charts. So far everything works fine if there are multiple bars shown, but I am getting empty graphs when I want to show only one bar.
So I created different plots with some dummy data, that shows my problem. (The data is fictitious)
data <- data.table(FRUIT = c("apples", "bananas", "kiwis", "apples", "bananas", "kiwis"),
KCALS = c(100, 200, 130, 100, 200, 130),
DATE = as.Date(c("2022-10-01", "2022-10-01", "2022-10-01", "2022-10-02", "2022-10-03", "2022-10-03")))
The first graph shows an overview over multiple dates:
plot_ly(data, x = ~DATE, y = ~KCALS, color = ~FRUIT, type = "bar") %>%
layout(yaxis = list(title = " "), xaxis = list(title = " "), barmode = "stack")
The second graph shows only one day:
plot_ly(data[DATE == "2022-10-01"], x = ~DATE, y = ~KCALS, type = "bar") %>%
layout(yaxis = list(title = " "), xaxis = list(title = " "), barmode = "stack")
Ok, that worked. But if I enable the "color-option" again, the plot is empty:
plot_ly(data[DATE == "2022-10-01"], x = ~DATE, y = ~KCALS, color = ~FRUIT, type = "bar") %>%
layout(yaxis = list(title = " "), xaxis = list(title = " "), barmode = "stack")
Why is it behaving like this? What am I doing wrong? And is there any documentation available, that really describes how the data gets processed in the plot_ly()- function?

When I run your code with color variable it does return a stacked bar plot for a single value. Make sure you don't overwrite variables and try it with a clean environment. Here is some reproducible code:
library(data.table)
data <- data.table(FRUIT = c("apples", "bananas", "kiwis", "apples", "bananas", "kiwis"),
KCALS = c(100, 200, 130, 100, 200, 130),
DATE = as.Date(c("2022-10-01", "2022-10-01", "2022-10-01", "2022-10-02", "2022-10-03", "2022-10-03")))
library(plotly)
plot_ly(data, x = ~DATE, y = ~KCALS, color = ~FRUIT, type = "bar") %>%
layout(yaxis = list(title = " "), xaxis = list(title = " "), barmode = "stack")
plot_ly(data[DATE == "2022-10-01"], x = ~DATE, y = ~KCALS, type = "bar") %>%
layout(yaxis = list(title = " "), xaxis = list(title = " "), barmode = "stack")
plot_ly(data[DATE == "2022-10-01"], x = ~DATE, y = ~KCALS, color = ~FRUIT, type = "bar") %>%
layout(yaxis = list(title = " "), xaxis = list(title = " "), barmode = "stack")
Created on 2023-02-16 with reprex v2.0.2
Here you can find the docs about bar charts in plotly.

How can I round decimals in plotly?

Is there a way to round decimals in plotly? For example,
I am getting the values as 24.32544 M.
Instead of this, can I get as 24.32 M?
asd <- data.frame(a = c(1,2,3,4), b = c(24325443,35345442,3245353453,345353523), c = c(5435352345,234534534,324534534,23453532))
plot_ly(asd, x = ~a, y = ~b, name = 'b', type = 'scatter', mode = 'lines') %>%
add_trace(y = ~c, name = 'c', mode = 'lines') %>%
layout(xaxis = list(title = paste0("Week"),showgrid = F,rangemode = "tozero"),
yaxis = list(title = "",showgrid = F,rangemode = "tozero"),
legend = legend_cus,
hovermode = 'x unified')

p <- ggplot(asd, aes(a, b,
text = paste0("a: ", round(a), "</br></br>b: ", round(b))))
+ geom_point()
ggplotly(p, tooltip = "text")
To conver to million format see: Format numbers with million (M) and billion (B) suffixes

Plotly: dual y axis graph messing up line graph

I'm trying to make a dual axis plot of rainfall and temperature. I have ordered the months on the bottom, but that causes my line graph to screw up. How do I make sure the added line uses the same x axis?
temprain<-data.frame(month = c(1:12),
Train = c(250,220, 180,97,38,27,31,47,70,140,200,250),
Tair = c(17,16, 15,13,9,6,5,9,12,13,14,16))
tempseq<-seq(0,20,by=0.5)
rainseq<-seq(0,260,by=1)
xlab<-list(type = "category",
categoryorder = "array",
categoryarray = month.name,
showgrid = TRUE,
showline = TRUE,
autorange = TRUE,
showticklabels = TRUE,
ticks = "outside",
tickangle = 0
)
plot_ly(temprain) %>%
add_bars(x = ~MonthName, y = ~Train, type = "bar", name = "Rain") %>%
add_lines(x = ~MonthName, y = ~Tair, yaxis = "y2", name = "Temp") %>%
layout(xaxis = xlab,
yaxis = list(showline = TRUE, side = "left",
title = "Rainfall (mm)Temp", range = tempseq),
yaxis2 = list(showline = TRUE, side = "right",
overlaying = "y", title = "Air Temp (C)", range = rainseq),
showlegend = FALSE,
margin = list(pad = 0, b = 50, l = 50, r = 50))
I tried this as well, and it doesn't work, the temp graph disappears
plot_ly(temprain, x = ~MonthName, y = ~Tair, name = "Temp") %>%
add_bars(x = ~MonthName, y = ~Train, yaxis = "y2", type = "bar", name = "Rain") %>%
layout(xaxis = xlab,
yaxis = list(showline = TRUE, side = "left",
title = "Air Temp (C)", range = tempseq),
yaxis2 = list(showline = TRUE, side = "right",
overlaying = "y",
title = "Rainfall (mm)", range = rainseq),
showlegend = FALSE,
margin = list(pad = 0, b = 50, l = 50, r = 50))

Below is the solution:
Your data:
temprain<-data.frame(month = c(1:12),
Train = c(250,220, 180,97,38,27,31,47,70,140,200,250),
Tair = c(17,16, 15,13,9,6,5,9,12,13,14,16))
Generate a column for month abbreviations from month:
mymonths <- c("Jan","Feb","Mar",
"Apr","May","Jun",
"Jul","Aug","Sep",
"Oct","Nov","Dec")
# match the month numbers against abbreviations:
temprain$MonthAbb = mymonths[ temprain$month ]
# This is the code to archieving a consistent combined graph:
temprain$MonthAbb <- factor(temprain$MonthAbb, levels = c(as.character(temprain$MonthAbb)))
Now plot your data:
fig <- plot_ly(temprain)
# Add the Train trace:
fig <- fig %>% add_trace(x = ~MonthAbb, y = ~Train, name = "Train", type = "bar")
ay <- list(
tickfont = list(color = "red"),
overlaying = "y",
side = "right",
title = "<b>Tair</b>")
# Add the Tair trace:
fig <- fig %>% add_trace(x = ~MonthAbb, y = ~Tair, name = "Tair", yaxis = "y2", mode = "lines+markers", type = "scatter")
fig <- fig %>% layout(yaxis2 = ay,
xaxis = list(title="Month"),
yaxis = list(title="<b>Train</b>"))%>%
layout(xaxis = list(
zerolinecolor = '#ffff',
zerolinewidth = 2,
gridcolor = 'ffff'),
yaxis = list(
zerolinecolor = '#ffff',
zerolinewidth = 2,
gridcolor = 'ffff')
)
fig
Output:

Switch displayed traces via plotly dropdown menu

I am working with the R programming language. I am trying to replicate this tutorial over here for my own data: https://plotly.com/r/dropdowns/
I created some fake data and made 4 plots:
#load libraries
library(plotly)
library(MASS)
library(dplyr)
# create data
x <- sample( LETTERS[1:4], 731, replace=TRUE, prob=c(0.25, 0.25, 0.25, 0.25) )
y <- rnorm(731,10,10)
z <- rnorm(731,5,5)
date= seq(as.Date("2014/1/1"), as.Date("2016/1/1"),by="day")
df <- data.frame(x,y, z, date)
df$x = as.factor(df$x)
# plot 1 : time series
aggregate = df %>%
mutate(date = as.Date(date)) %>%
group_by(month = format(date, "%Y-%m")) %>%
summarise( mean = mean(y))
ts_1 <- ggplot(aggregate) + geom_line(aes(x = month, y = mean, group = 1)) + theme(axis.text.x = element_text(angle = 90)) + ggtitle("time series 1")
plot_1 = ggplotly(ts_1)
#plot 2 : box plot
plot_2 <- plot_ly(df, y = ~y, color = ~x, type = "box") %>% layout(title = "boxplot")
#plot 3, 4 : scatter plots
df_1 <- df[which(df$x == "A"),]
df_2 <- df[which(df$x == "B"),]
plot_3 <- plot_ly( data = df_1, type = "scatter", mode = "markers", x = ~ y, y = ~z) %>% layout(title = "graph 3")
plot_4 <- plot_ly( data = df_2, type = "scatter", mode = "markers", x = ~ y, y = ~z) %>% layout(title = "graph 4")
Once these 4 plots have been created, I know how to save them together:
sub = subplot(plot_1, plot_2, plot_3, plot_4, nrows = 2)
#view result
sub
Now what I am trying to do, is have the user "toggle" (switch) between these graphs (as seen here: https://plotly.com/r/dropdowns/)
In a previous post (R: Switching Between Graphs ), I learned how to "glue" similar graphs together (e.g. 4 scatter plots). Now, I am trying to do so with different graphs (2 scatter plots, 1 time series and 1 box plot). I tried to adapt the code from the previous post to suit my example:
fig <- df %>%
add_trace(name = "A", plot_1) %>%
add_trace (name = "B" , df, y = ~y, color = ~x, type = "box") %>% layout(title = "boxplot")
add_trace (name = "C" , data = df_1, type = "scatter", mode = "markers", x = ~ y, y = ~z) %>% layout(title = "graph 3") %>%
add_trace( name = "D", data = df_2, type = "scatter", mode = "markers", x = ~ y, y = ~z) %>% layout(title = "graph 4") %>%
layout(xaxis = list(domain = c(0.1, 1)),
yaxis = list(title = "y"),
updatemenus = list(
list(
y = 0.7,
buttons = list(
list(method = "restyle",
args = list("visible", list(TRUE, FALSE, FALSE, FALSE)),
label = "A"),
list(method = "restyle",
args = list("visible", list(FALSE, TRUE, FALSE, FALSE)),
label = "B"),
list(method = "restyle",
args = list("visible", list(FALSE, FALSE, TRUE, FALSE)),
label = "C"),
list(method = "restyle",
args = list("visible", list(FALSE, FALSE, FALSE, TRUE)),
label = "D")))))
But this produces the following errors:
Error: $ operator is invalid for atomic vectors
Error in add_data(p, data) : argument "p" is missing, with no default
Can someone please show me if it is possible to fix this problem? Instead of using the "add_trace" approach, is it somehow possible to individually call each plotly graph object by its name (e.g. subplot(plot_1, plot_2, plot_3, plot_4, nrows = 2)), "glue" all the graphs together, and then add a "toggle button" that lets the user switch between them?
(note: I need to be able to save the final result as a "html" file)
Thanks

First of all, you should take care about plots which add multiple traces (see nTracesA etc.)
Besides changing the trace visibility you'll need to seperate categorial and numerical data onto separate x and y-axes and manage their visibility, too (see xaxis2, xaxis3, xaxis4 - this also works with a single y-axis but in this case the grid isn't displayed properly)
As described in the docs:
The updatemenu method determines which plotly.js function will be used
to modify the chart. There are 4 possible methods:
"restyle": modify data or data attributes
"relayout": modify layout attributes
"update": modify data and layout attributes
"animate": start or pause an animation (only available offline)
Accordingly the following, is using the update method (a lot of repition here - needs some cleanup, but I think it's better to understand this way):
# load libraries
library(dplyr)
library(plotly)
# create data
x <- sample(LETTERS[1:4],
731,
replace = TRUE,
prob = c(0.25, 0.25, 0.25, 0.25))
y <- rnorm(731, 10, 10)
z <- rnorm(731, 5, 5)
date <- seq(as.Date("2014/1/1"), as.Date("2016/1/1"), by = "day")
df <- data.frame(x, y, z, date)
df$x = as.factor(df$x)
nTracesA <- nTracesC <- nTracesD <- 1
nTracesB <- length(unique(df$x))
plotA <- plot_ly(data = df %>%
mutate(date = as.Date(date)) %>%
group_by(month = format(date, "%Y-%m")) %>%
summarise(mean = mean(y)),
type = 'scatter', mode = 'lines', x= ~ month, y= ~ mean, name = "plotA", visible = TRUE, xaxis = "x", yaxis = "y")
plotAB <- add_trace(plotA, data = df, x = ~x, y = ~y, color = ~ x, name = ~ paste0("plotB_", x),
type = "box", xaxis = "x2", yaxis = "y2", visible = FALSE, inherit = FALSE)
plotABC <- add_trace(plotAB, data = df[which(df$x == "A"),],
type = "scatter", mode = "markers", x = ~ y, y = ~ z,
name = "plotC", xaxis = "x3", yaxis = "y3", visible = FALSE, inherit = FALSE)
plotABCD <- add_trace(plotABC, data = df[which(df$x == "B"),], x = ~ y, y = ~ z,
type = "scatter", mode = "markers", name = "plotD", xaxis = "x4", yaxis = "y4", visible = FALSE, inherit = FALSE)
fig <- layout(plotABCD, title = "Initial Title",
xaxis = list(domain = c(0.1, 1), visible = TRUE, type = "date"),
xaxis2 = list(overlaying = "x", visible = FALSE),
xaxis3 = list(overlaying = "x", visible = FALSE),
xaxis4 = list(overlaying = "x", visible = FALSE),
yaxis = list(title = "y"),
yaxis2 = list(overlaying = "y", visible = FALSE),
yaxis3 = list(overlaying = "y", visible = FALSE),
yaxis4 = list(overlaying = "y", visible = FALSE),
updatemenus = list(
list(
y = 0.7,
buttons = list(
list(label = "A",
method = "update",
args = list(list(name = paste0("new_trace_name_", 1:7), visible = unlist(Map(rep, x = c(TRUE, FALSE, FALSE, FALSE), each = c(nTracesA, nTracesB, nTracesC, nTracesD)))),
list(title = "title A",
xaxis = list(visible = TRUE),
xaxis2 = list(overlaying = "x", visible = FALSE),
xaxis3 = list(overlaying = "x", visible = FALSE),
xaxis4 = list(overlaying = "x", visible = FALSE),
yaxis = list(visible = TRUE),
yaxis2 = list(overlaying = "y", visible = FALSE),
yaxis3 = list(overlaying = "y", visible = FALSE),
yaxis4 = list(overlaying = "y", visible = FALSE)))
),
list(label = "B",
method = "update",
args = list(list(visible = unlist(Map(rep, x = c(FALSE, TRUE, FALSE, FALSE), each = c(nTracesA, nTracesB, nTracesC, nTracesD)))),
list(title = "title B",
xaxis = list(visible = FALSE),
xaxis2 = list(overlaying = "x", visible = TRUE),
xaxis3 = list(overlaying = "x", visible = FALSE),
xaxis4 = list(overlaying = "x", visible = FALSE),
yaxis = list(visible = FALSE),
yaxis2 = list(overlaying = "y", visible = TRUE),
yaxis3 = list(overlaying = "y", visible = FALSE),
yaxis4 = list(overlaying = "y", visible = FALSE)))),
list(label = "C",
method = "update",
args = list(list(visible = unlist(Map(rep, x = c(FALSE, FALSE, TRUE, FALSE), each = c(nTracesA, nTracesB, nTracesC, nTracesD)))),
list(title = "title C",
xaxis = list(visible = FALSE),
xaxis2 = list(overlaying = "x", visible = FALSE),
xaxis3 = list(overlaying = "x", visible = TRUE),
xaxis4 = list(overlaying = "x", visible = FALSE),
yaxis = list(visible = FALSE),
yaxis2 = list(overlaying = "y", visible = FALSE),
yaxis3 = list(overlaying = "y", visible = TRUE),
yaxis4 = list(overlaying = "y", visible = FALSE)))),
list(label = "D",
method = "update",
args = list(list(visible = unlist(Map(rep, x = c(FALSE, FALSE, FALSE, TRUE), each = c(nTracesA, nTracesB, nTracesC, nTracesD)))),
list(title = "title D",
xaxis = list(visible = FALSE),
xaxis2 = list(overlaying = "x", visible = FALSE),
xaxis3 = list(overlaying = "x", visible = FALSE),
xaxis4 = list(overlaying = "x", visible = TRUE),
yaxis = list(visible = FALSE),
yaxis2 = list(overlaying = "y", visible = FALSE),
yaxis3 = list(overlaying = "y", visible = FALSE),
yaxis4 = list(overlaying = "y", visible = TRUE))))
))))
print(fig)
# htmlwidgets::saveWidget(partial_bundle(fig), file = "fig.html", selfcontained = TRUE)
# utils::browseURL("fig.html")
Some related info:
https://plotly.com/r/custom-buttons/
https://plotly.com/r/multiple-axes/

This is just a guess from the documentation but there is no add_data() call so maybe try this for your first line:
fig <- plot_ly() %>% add_data(df) %>%
See docs example:
plot_ly() %>% add_data(economics) %>% add_trace(x = ~date, y = ~pce)

A user on the Rstudio community forum provided an answer : https://community.rstudio.com/t/gluing-graphs-together-switch-toggle-between-graphs-in-r-plotly/95891/3
I am still trying to figure out how to format the axis - maybe someone could take a look at this?
#load libraries
library(plotly)
library(MASS)
library(dplyr)
# create data
x <- sample( LETTERS[1:4], 731, replace=TRUE, prob=c(0.25, 0.25, 0.25, 0.25) )
y <- rnorm(731,10,10)
z <- rnorm(731,5,5)
date= seq(as.Date("2014/1/1"), as.Date("2016/1/1"),by="day")
df <- data.frame(x,y, z, date)
df$x = as.factor(df$x)
# plot 1 : time series
aggregate = df %>%
mutate(date = as.Date(date)) %>%
group_by(month = format(date, "%Y-%m")) %>%
summarise( mean = mean(y))
ts_1 <- ggplot(aggregate) + geom_line(aes(x = month, y = mean, group = 1)) + theme(axis.text.x = element_text(angle = 90)) + ggtitle("time series 1")
plot_1 = ggplotly(ts_1)
#plot 2 : box plot
plot_2 <- plot_ly(df, y = ~y, color = ~x, type = "box") %>% layout(title = "boxplot")
#plot 3, 4 : scatter plots
df_1 <- df[which(df$x == "A"),]
df_2 <- df[which(df$x == "B"),]
plot_3 <- plot_ly( data = df_1, type = "scatter", mode = "markers", x = ~ y, y = ~z) %>% layout(title = "graph 3")
plot_4 <- plot_ly( data = df_2, type = "scatter", mode = "markers", x = ~ y, y = ~z) %>% layout(title = "graph 4")
fig = plot_ly()
fig = fig %>% add_trace(data = df %>%
mutate(date = as.Date(date)) %>%
group_by(month = format(date, "%Y-%m")) %>%
summarise( mean = mean(y)), type = 'scatter', mode = 'lines', x= ~month, y= ~mean,
name = "timeseries")
fig = fig %>% add_trace(data = df[which(df$x == "A"),], y = ~y, color = ~x,
type = "box", name = "boxplot")
fig = fig %>% add_trace( data = df[which(df$x == "B"),],
type = "scatter", mode = "markers", x = ~ y, y = ~z,
name= "graph2")
fig = fig %>% add_trace(data = df[which(df$x == "A"),], y = ~y, color = ~x,
type = "box", name = "boxplot2")
fig %>% layout(xaxis = list(domain = c(0.1, 1)),
yaxis = list(title = "y"),
updatemenus = list(
list(
y = 0.7,
buttons = list(
list(method = "restyle",
args = list("visible", list(TRUE, FALSE, FALSE, FALSE)),
label = "A"),
list(method = "restyle",
args = list("visible", list(FALSE, TRUE, FALSE, FALSE)),
label = "B"),
list(method = "restyle",
args = list("visible", list(FALSE, FALSE, TRUE, FALSE)),
label = "C"),
list(method = "restyle",
args = list("visible", list(FALSE, FALSE, FALSE, TRUE)),
label = "D")))))

Localize plotly plot time axis for dates in R

I would like to localize a plotly graph date axis date format for a special country. Instead of labeling dates in the time axis as "Jan 8 Jan 9 Jan 10...", I want to label them "Oca 8 Oca 9 Oca 10" in Turkish locale. The sample plotly code in R is given below. I know that I need to use config(locale = "tr_TR") to do this. But I am not sure how to do this in the code below.
plotly::plot_ly() %>%
plotly::add_trace(
x = ~date,
y = ~Belgium,
type = "scatter",
mode = "lines+markers",
name = "Türkiye"
) %>%
plotly::add_trace(
x = ~date,
y = ~France,
type = "scatter",
mode = "lines+markers",
name = "Fransa"
) %>%
plotly::add_trace(
x = ~date,
y = ~Spain,
type = "scatter",
mode = "lines+markers",
name = "İspanya"
) %>%
plotly::add_trace(
x = ~date,
y = ~Italy,
type = "scatter",
mode = "lines+markers",
name = "Italya"
) %>%
plotly::layout(
title = "",
legend = list(x = 0.1, y = 0.9),
yaxis = list(title = "Toplam Onaylı Vakalar"),
xaxis = list(title = "Date"),
# paper_bgcolor = "black",
# plot_bgcolor = "black",
# font = list(color = 'white'),
hovermode = "compare",
margin = list(
# l = 60,
# r = 40,
b = 10,
t = 10,
pad = 2
)
)```

I just added the following
)%>%
plotly::config(
locale='tr')
)
This configuration sets the date axis labeling based on system locale.

Resources