The Weekly Challenge ‐ Perl and Raku

CY's Take on The Weekly Challenge #165

If you want to challenge yourself on programming, especially on Perl and/or Raku, go to https://theweeklychallenge.org, code the latest challenges, submit codes on-time (by GitHub or email).

Do tell me, if I am wrong or you strongly oppose my statements!

It's time for challenges in Week #165 !

Task 1: Scalable Vector Graphics (SVG)

The task is beginner-friendly and introduces us to the CPAN module SVG. I did some prioritization so the image might look better: putting the drawing of dots in the later stage.

my $img = SVG->new;
my @points;

while (<>) {
    chomp;
    my @nums = split ",", $_;
    if (scalar @nums == 2) {
        push @points, [@nums];
    }
    elsif (scalar @nums == 4) {
        new_line(@nums);
    }
}

new_point($_->@*) foreach @points; 

This week I also made use of the CPAN module ARGV::OrDATA by teammate Mr Choroba, so the testing can be shown more easily.

a random plot

Task 2: Line of Best Fit

I am not good at statistics. Just followed the instructions on the mathisfun.com link provided.

my $n = scalar @points;

my $slope =
     (  $n * sum(map {$_->[0]*$_->[1]} @points)  # n sum(xy) - sum(x) sum(y)
      - sum(map{$_->[0]} @points) * sum(map{$_->[1]} @points) )
    / ( $n * sum(map{$_->[0]**2} @points)        # n sum(sq x) - sq(sum x)
         - (sum( map { $_->[0]} @points ))**2) ;


my $intercept =
     ( sum( map {$_->[1]} @points )             # sum(y) - slope * sum(x)
    - $slope * sum( map {$_->[0]} @points) )
    / $n ;

sub y_best_fit {
    my $x = $_[0];
    return $slope*$x + $intercept;
}

Another skill I am bad at is packaging. So an extra data file is created and then the Task 2 script forcefully calls the command perl ch-1.pl < BEST_FIT_DATA > new.svg to complete the Task 2 via script in Task 1.

my $line_of_best_fit =
    join ",", $minx,y_best_fit($minx),$maxx,y_best_fit($maxx);

open FH, "> BEST_FIT_DATA" or die $!;
say FH $line_of_best_fit;
say FH $_->[0],",",$_->[1] for @points;

system("perl ch-1.pl < BEST_FIT_DATA > new.svg"); # applicable on Linux

After finishing the tasks I recalled Julia. Julia is famous for scientific applications and the least square fit is a basic tool in statistics. After some search I encountered this blogpost: "Simplest least-squares in Julia", by Jacob Louis Hoover (11th Jan 2021). I have followed it step-by-step except the operator "\" does not work magically in the current version of Julia. Anyway we can touch more mathematics looking at the task via the lens of matrix operations (See also Moore-Penrose pseudoinverse.)

The image by Julia is different from that by the Task 2 Perl code and SVG module! Calm down and look patiently: the Plots in Julia counts the y-coordinates from bottom to top, while the SVG module interpreted them from top to bottom. It would be more obvious if we emphasize to the Julia interpreter to plot the graph with x-y ratio equal to 1:

plot!(
    x -> slope*x + intercept,
    legend = :none,
    ratio = 1,               #optional
)

Stay alert and healthy! □

Except from images and codes from other personnels, the content of this blogpost is released under a copyleft spirit. One may share (full or partial) content of this blogpost on other platform if you share it under the free and open content spirit.

link for CY's full codes: ch-1.pl, ch-2.pl, ch-2.jl

Contact on twitter: @e7_87.

Discuss via GitHub issues: here.

Email: fungcheokyin at gmail.com

Created Date: 17th May, 2022.

Published Date: 20th May, 2022.

Footnote: After releasing the blogpost, I realize that some points on the boundary via my Task 1 script are partially cut. -- Friday, May 20, 2022 PM09:45:00 HKT