Ancestors

Toot

Written by Jordi Rosell on 2024-12-13 at 09:59

I used F1 score in the past in many machine learning projects in #rstats and #python (It's useful for imbalanced classification problems), but I hadn't visualized its relationship with precision and recall UNTIL TODAY :)

Here a {rayshader} #dataviz comparing F1 score harmonic mean (As you see, it penalizes the score if one metric is bad) with the mean (It simply averages metrics).

=> View attached media | View attached media

=> More informations about this toot | More toots from jrosell@mastodon.social

Descendants

Written by Jordi Rosell on 2024-12-13 at 10:02

Thanks to @koaning for pointing out this visualization in Stack Overflow https://stackoverflow.com/a/49535509/481463 in this video https://www.youtube.com/watch?v=3M2Gmuh5mtI

=> More informations about this toot | More toots from jrosell@mastodon.social

Written by Jordi Rosell on 2024-12-13 at 10:04

Here the code: https://gist.github.com/jrosell/667fee58c8ff47a99591fa6122e04ed3

=> More informations about this toot | More toots from jrosell@mastodon.social

Written by Mario Angst on 2024-12-13 at 10:56

@jrosell really cool viz!

=> More informations about this toot | More toots from mario_angst_sci@fediscience.org