MNIST binary classification tutorial

[BrunoLiegiBastonLiegi]

As per title.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 66.59%. Comparing base (807c89b) to head (5b77072).
Report is 16 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main      #49   +/-   ##
=======================================
  Coverage   66.59%   66.59%           
=======================================
  Files          11       11           
  Lines         488      488           
=======================================
  Hits          325      325           
  Misses        163      163           

Flag	Coverage Δ
unittests	66.59% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

---- 🚨 Try these New Features:

• Flaky Tests Detection - Detect and resolve failed and flaky tests

[BrunoLiegiBastonLiegi]

@MatteoRobbiati @niccololaurora can you double check that the notebook is clear?

[MatteoRobbiati]

Some small comments:

1. I would add subtitles, it will lighten the text (which is very clear but some blocks of text would benefit of subtitles);
2. I would put as optional the installation of the packages (namely, comment the first cell or something equivalent);
3. Add labels to $x$ and $y$ axis of loss function plot;
4. Introduce the F1 score or link to some reference;
5. not necessary - I would add the same image of the first code cell with the mnist digit and the new prediction. In the future I will write a function to plot e.g. 10 images with titles in red and green showing misclassification.

[BrunoLiegiBastonLiegi]

Thanks, I should have addressed everything.