With the help of LLMs, I have partially solved a long-troublesome issue with GBDTs!
1. Early stopping is a nice, clean way to automatically tune your GBDT
2. You by default leave some performance on the table with your hold-out set ...
@lymanstoneky the 17% reduction in mortality is deaths from all causes, not malaria deaths, so the drop in malaria deaths is likely much larger than that
OpenRAPM now has "Wins over Starter Average"
Why: When thinking about MVP, who cares about 'replacement player'?
How: Wins Added vs the avg starter this season - reducing minute weight during garbage time, & accounting for player effect on pace
https://t.co/i06UvRdo2v
@yolo_pinyato@kmedved (sorry for reiterating what you already said to some degree)
but in my experience - if you just measure player change between RS/playoffs you moreso are measuring coaching changes, player improvement, player injury
Wolves Fans:
For the first time in 7 years of posting video clips, I received a copyright complaint from the NBA on the clips posted last night
If you have found @WolvesClips to be valuable to your NBA fandom, I’d really appreciate you sharing what this account has meant to you
@yolo_pinyato All good RAPM metrics account for this to some degree. In my testing this isn't really cut and dry. Would be curious to see what your testing showed