subplots throw away shape information

[ashleightrinh]

Is there a way to layer plots? So say I have a histogram, I want to be able to plot a line over it indicating the mean. Also maybe to have multiple histograms overlaid.

Thanks,
Ashleigh

[sglyon]

Hi @ashleightrinh!

yes this is possible.

You can just add multiple traces to the plot.

here's an example:

plot([scatter(y=50 .* rand(10)), histogram(x=10 .* rand(1000))])

[ashleightrinh]

Thanks Spencer, Though is there a way to plot just a vertical line (say, at the mean)? I have tried plotting just a point using scatter but nothing appears.

…

On 29 September 2017 at 15:16, Spencer Lyon ***@***.***> wrote: Hi @ashleightrinh <https://github.com/ashleightrinh>! yes this is possible. You can just add multiple traces to the plot. here's an example: plot([scatter(y=50 .* rand(10)), histogram(x=10 .* rand(1000))]) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#138 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AYCDWLCF2mOClZV450Ku2rIh56R5g_ywks5snPu_gaJpZM4Posw_> .

[sglyon]

Yep, check out the shapes docs: http://spencerlyon.com/PlotlyJS.jl/examples/shapes/

You are looking for using the vline function to add shapes to the layout

[asbisen]

below is a dirty sample

using PlotlyJS

x=collect(1:100)
y=rand(100)
m=mean(y)

t0=scatter(x=x, y=y)
s=[hline(m)]

plot([t0], Layout(;shapes=s))

[ashleightrinh]

Thank you! I realised it does work just not when I am trying to put multiple plots next to each other. Here is the code I have

shape1 = [vline(mean(matrix1))]
shape2 = [vline(mean(matrix2))]
shape3 = [vline(mean(matrix3))]

data1 = histogram(x=matrix1, nbinsx=18)
data2 = histogram(x=matrix2, nbinsx=18)
data3 = histogram(x=matrix3, nbinsx=18)

where <matrix1>, <matrix2> and <matrix3> are of type <Array{Float64,1}>.

p1 = plot([data1], Layout(;shapes=shape1))
p2 = plot([data2], Layout(;shapes=shape2))
p3 = plot([data3], Layout(;shapes=shape3))
[p1 p2 p3]

[sglyon]

I think you are right. I don't currently have logic to combine the shapes from individual plot layouts into the final "subplotted" plot.

You could try

p1 = plot([data1], Layout(;shapes=shape1))
p2 = plot([data2], Layout(;shapes=shape2))
p3 = plot([data3], Layout(;shapes=shape3))
p = [p1 p2 p3]
relayout!(p, shapes=vcat(shape1, shape2, shape3))

But I'm worried that the locations of the shapes will not be correct...

[sglyon]

Looks like all issues here have been resolved except the subplots + shapes one, making this a dup of #100. Closing in favor of that issue.