Add images data classification w/ SVM and squeeze out a lot of performance

@mikolajr Motivation:

• classical images analysis also very good
• connects later to NNs getting even more performance

See MNIST SVM solutions from Kaggle / rankings page.

Alternatively: celebrities faces dataset classification w/ SVM from scikit-learn (also in VanderPlass book)

In script about pipelines or details on classifiers e.g. as a summary exercise (dep. on if tunning needs some insights).

changed milestone to %Backlog

changed the description

changed milestone to %Workshop 4th edition (Sep-Oct, 2021)

changed the description

Note: optionally to be done last-minute, after !43 (merged) and !45 (merged) are merged, as a summary exercise for the classifiers overview instead of the beers (only if easy to run): e.g. MNIST digits already used earlier, but faces would be nicer (if work)

created merge request !49 (merged) to address this issue

mentioned in merge request !49 (merged)

mentioned in commit a4b1966c

closed with merge request !49 (merged)

assigned to @mikolajr

mentioned in commit 5578dd50