math is more beautiful when you can visualize it.
implemented gradient-boosted decision trees after a long time. Seeing the curve fitting the non-linear data after every iteration makes it super easy to understand and build intuition.
don’t run away from math. It just demands a bit more effort than usual. spend time coding, spend time learning, spend time understanding.
if you’re interested, my experimental repository containing all my prototypes is linked in the comments. I highly recommend having such a repository where you can store all your prototypes.