Plots 3: Scatter and Histogram

I thought of ending blogs about plots last time, but I did think about Scatter and Histogram plots. I feel they are important and I felt I need them to better explain some of the blogs I kinda a plan to write in the future. So here are scatter and histogram plots.

So let’s create a simple scatter, x-values as first argument, y-values as second, I have given color as red but its completely optional:

scatter_plot = scatter([1, 2 , 3,  4], [1, 2, 3, 4], color = "red")

Output:

We have captured last output in a variable called scatter_plot, this is much better variable name I tell you. In my previous blog I had given a names like p which is actually blasphemy in programming world, you might be smart today and understand what p is, but possibly 6 months later it will haunt you, or a programmer touching your code sometime later might curse you thus causing your loved ones to vomit blood and die. Anyway, we now add another scatter to scatter_plot using the code below:

scatter!(scatter_plot, [1, 7, 8, 2, 3], [2, 6, 3, 1, 4], color = "blue")

Output:

As you can see above, this time we have not used plot!() to modify scatter_plot but we have used this function scatter!() which I personally feel is much better to read. And we have put these new dots in color blue. Let’s modify scatter_plot again this time using plot!() function as shown below:

plot!(
  scatter_plot,
  [1, 2, 3, 4 ],
  [-1, -2 , -3, -4],
  color = "orange",
  seriestype = :scatter
)

Output:

As you see above in the above plot!() we have used named argument seriestype and have set it to :scatter for it to be a scatter plot. I am unsure why we have a colon : before scatter, should check Julia docs about it. Okay, looks like this is a special kind of thing known as Symbol, another data type in Julia, possibly it occupies less space compared to "scatter" which is a string when called multiple times. I just checked this code:

julia> :a
:a

julia> typeof(:a)
Symbol

just to check its type.

Okay, the label box int above scatter plots is overlapping a data point, lets increase the x right limit to 12 so that it would look better:

plot!(scatter_plot, xlims = (0, 12))

Output:

Better now!

For some reason I like histograms, so I have plotter a histogram below:

histogram(rand(1:1000, 500), bins = 20)

Output:

I think I will be using it while writing about the Iris data set. You can get the notebook file for this blog here https://gitlab.com/data-science-with-julia/code/-/blob/master/plots.ipynb.

Learn more about Plots

This is the last blog about Plots I think unless i change my mind for some reason, if you want to learn more check the official Plots website https://docs.juliaplots.org/latest/, and this Julia Plots by Prude University https://www.math.purdue.edu/~allen450/Plotting-Tutorial.html is good too.