Boxenplot - Seaborn

preview_player
Показать описание
This video is sort of an addendum to Seaborn Understanding the Weird Parts (since this came out a little bit after I recorded the rest of the Seaborn series). Here I go over boxenplot (which is pretty similar to boxplot, but is a LOT more useful!) and show you what it can do and why it is so useful! Enjoy!
Рекомендации по теме
Комментарии
Автор

Things I learned from this lecture:
Boxenplot: much more useful than boxplot for the data that is long-tailed

arhataria
Автор

thanks a lot for recording this! helped me verify that the inner-most box represents 25-75% IQR and the one to the right of it represents the 75-87.5% IQR. Useful plot to evaluate our right-tailed absolute model error distribution without relying on rather arbitrary box plot whiskers.

FelixLast
Автор

Tks a lot for this video! You're awesome!

manoelcamilodesousa
Автор

they said: IQR the data!
me: Long tails?
they said: Winsorize!
me: Still long tails!?
they said: RobustScaler! QuantileTransformer! SciKit the data!
me: Throw the data away?
they: ...You're new here, aren't you?
me: Yes 😭
YOU: Boxenplot to save the data!
me: THANK YOU

itsreallysimple
Автор

What is the advantage of this over just plotting the frequency histogram with the median overlaid, and percentiles on the x axis?

frederiktheissen
Автор

Thank you so much! Boxenplot is all too useful for Data Scientists..

nikhilbalwani