Second answer – maybe I was not clear… What I meant is: consider a surface plot of the Rosenbrock function:
function rosenbrock(x,y)
return (1-x)^2 + 100*(y-x^2)^2
end
x = linspace(-2,2)
y = linspace(-1,3)
plot(x,y,rosenbrock,st=:surface)
a: I want to add contour lines in the plot (that used to be possible, but I cannot find the keyword now (i.e., instead of “:surface”)
b: Often, the surface + contour plots don’t show off so well – typically when the surface hides the contours. So – how can I offset the z-position of the contours from z=0 to, say, z = -500?
“Concerning layout, this tutorial (3) may be helpful.”
My question was related to this tutorial, I guess… which says:
"The @layout macro is the easiest way to define complex layouts, using Julia’s multidimensional Array construction as the basis for a custom layout syntax. Precise sizing can be acheived with curly brackets, otherwise the free space is equally split between the plot areas of subplots.
However, the tutorial does not state the meaning of “a” and “b” in the constructs " a{0.3w}" and “b{0.2h}”. My question is on what these mean. [I can guess, but perhaps it should not be necessary to guess??]
OK – I’ll take a look at it. Personally, I like to print a pdf manual over using a web based tool; to me, that helps in learning. But others may feel differently. I will look over my private notes on Matplotlib/PyPlot from when I focused on Python, and see how my experience from that period can be useful in the Plots documentation structure. It will take a couple of weeks to summarize my thoughts, though.
Since you are asking me to advise, I will :-). Don’t use jet! It doesn’t have good properties for scientific visualization. Use a colorscheme that has.
But to answer your question, you do cgrad(:jet, reverse = true) today. The docs are massively improved today, btw ColorSchemes · Plots