Plotting multiple traces on a single plot in IJulia/Jupyter using a loop

General Usage
1

Does any know how to add multiple traces to a plot in Jupyter?

The following code works fine in Juno. It generates a series of offset sine waves in the same plot.

using GR

x=collect(0:0.1:2pi)
y=sin.(x)

plot(x,y)

for i=1:10
oplot(x,y.+i/10)
end

However, in Jupyter, no plot is produced. It seems anytime a cell ends in something other than the direct call for a plot, the plot does not show up when the cell is run.

Running Julia 1.0.3

2

Found a solution using a function. Not sure why the original does not work (maybe a scope related issue?).

Correction: the code below only worked once by accident, not sure what I did. For the correct implementation see Simon_Bolland’s response below

using GR

x=collect(0:0.1:2pi)
y=sin.(x)

function tryplot()
plot(x,y)
for i=1:10
    z=oplot(x,y.+i/10)
end
    return z
end

tryplot()

Still, if I try this code in Jupyter:

using GR

x=collect(0:0.1:2pi)
y=sin.(x)

plot(x,y)
println("Hello")

The only output from running the cell is “Hello”. If anyone has any insight on why no plot shows up in the latter case it would be appreciated.

3

Jupyter defaults to only showing the last value in a cell

Try

using GR

x=collect(0:0.1:2pi)
y=sin.(x)

display(plot(x,y))
println("Hello")

to force the plot to be shown

1 Like
4

Awesome. Thanks!. Now, any idea how to make that work in a loop so that all traces are plotting in the same plot?

using GR

x=collect(0:0.1:2pi)
y=sin.(x)

display(plot(x,y))
for i=1:10
    display(oplot(x,y.+i/10))
end
println("Hello")

The above generates 10 plots. All I need is the last one.

5 
using GR

x=collect(0:0.1:2pi)
y=sin.(x)
p=nothing 
for i=1:10
    p=oplot(x,y.+i/10)
end
display(p)
println("Hello")
1 Like
6

Excellent! Is the line below storing an array of svg images in p? If not, is there a way to do that so that the array elements (svg images) can be referenced/used elsewhere?

p=oplot(x,y.+i/10)
7

This overwrites p with the plot so far on each loop. If you want all the intermediate plots you could set up an array and do p[i]=oplot(x,y.+i/10) but I’ve shut pc down for the night so can’t look for correct element type - let me know if you need this and I’ll look tomorrow

8

I’m not finding the element type so any input you can provide tomorrow would be appreciated. Thanks!

9

Try this:

using GR

x=collect(0:0.1:2pi)
y=sin.(x)

p=Array{GR.SVG}(undef,1,11)
p[1]=plot(x,y)

for i=1:10
    p[i+1]=oplot(x,y.+i/10)
end
display(p[1])
display(p[6])
display(p[11])
println("Hello")
1 Like
10

Works great. Many thanks!

11

A quick addition for the benefit of anyone reading this later. A plot statement seems to be needed before the loop so the oplot function has an image to write over.

using GR

x=collect(0:0.1:2pi)
y=sin.(x)
p=nothing 
plot(x,y)    #<===============
for i=1:10
    p=oplot(x,y.+i/10)
end
display(p)
println("Hello")