Plotly : Difference between mode and add_trace

Question

dat = c(1:100)

fig1 = plot_ly(x = ~x)

fig2 = fig1%>%add_trace(y=~rnorm(100), mode= "lines")

The outputs for "fig1" and "fig2" are: and respectively.

I could also get "fig2" using the following code:

fig3 = plot_ly(x= ~dat, y = ~rnorm(100))%>%add_lines()

Can someone explain what's the difference individual traces like (add_histogram, add_lines etc..) and the function add_trace with an appropriate mode?

Answer 1

add_histogram , add_lines etc. are convenience functions with a preset trace type (internally they are setting the according type and call add_trace_classed - type eg add_lines in the console to inspect the function).

add_trace is a general function to creates traces of any available type .

Your third option to create traces is via the plot_ly() function itself.

Please see the example section of ?add_trace for additinal information:

the plot_ly() function initiates an object, and if no trace type is specified, it sets a sensible default p <- plot_ly(economics, x = ~date, y = ~uempmed) p

some add_*() functions are a specific case of a trace type for example, add_markers() is a scatter trace with mode of markers add_markers(p)

If you don't specify a trace type in plot_ly it is set based on the data you provided:

library(plotly)

dat = data.frame(x = 1:100)

fig1 = plot_ly(data = dat, x = ~x)
# No trace type specified:
#   Based on info supplied, a 'histogram' trace seems appropriate.
# Read more about this trace type -> https://plotly.com/r/reference/#histogram

fig1a = plot_ly(data = dat, x = ~x, type = "scatter", mode = "lines")

fig2 = fig1 %>% add_trace(y=~rnorm(100), mode = "lines")