Ah yes, regression
Submitted 6 months ago by The_Picard_Maneuver@lemmy.world to science_memes@mander.xyz
https://lemmy.world/pictrs/image/7b2dc821-73fe-45ab-8bf6-359c5aee30f8.jpeg
Comments
frezik@midwest.social 6 months ago
Corr@lemm.ee 6 months ago
That was a joy. Thank you for sharing
TachyonTele@lemm.ee 6 months ago
Check this shit out (fig 1).
Lmao there’s so much gold. I got frustrated for him while reading it.
credo@lemmy.world 6 months ago
I had a hard time reading. Could not stop laughing enough to get past the fourth paragraph. Sides hurt, would not recommend.
TheOakTree@lemm.ee 6 months ago
Germanium My Ass
CodexArcanum@lemmy.world 6 months ago
Going into physics was the biggest mistake of my life. I should’ve declared CS. I still wouldn’t have any women, but at least I’d be rolling in cash.
Honestly, there wasn’t all that much cash to roll in and there’s less all the time now. Plus, if you think busted equipment is bad, wait until I tell you about inheriting legacy code.
frezik@midwest.social 6 months ago
From what I heard, the guy actually did change to a CS major shortly after writing this.
ornery_chemist@mander.xyz 6 months ago
This relation between temperature and resistivity can be shown to be exponential in certain temperature regimes by waving your hands and chanting “to first order.”
for some reason this is the line that got me
negativenull@lemmy.world 6 months ago
One Line to rule them all
One Line to find them
One Line to bring them all
and in the data bind them_stranger_@lemmy.world 6 months ago
Aurenkin@sh.itjust.works 6 months ago
One line best-fits all
merari42@lemmy.world 6 months ago
Machine Learning enthusiasts: Why settle for linear regression when you can deploy a Gradient-Boosted Random Deep Neural Net Surface Vector Cluster that consumes the entire power of Iceland to trace a perfect ∞-dimensional hypersphere around blue points? Overparameterization is the future!
reinei@lemmy.world 6 months ago
Fine! I’ll use a second order polynomial to fit this instead. But that’s the last order I’m willing to go to!
prex@aussie.zone 6 months ago
Always XKCD: https://xkcd.com/882/
badbytes@lemmy.world 6 months ago
Oh, it’s trending up. That’s progress!
dream_weasel@sh.itjust.works 6 months ago
R^2=0.03
Engywuck@lemm.ee 6 months ago
Well, some students of mines actually put some figure like that in al lab report…
Whelks_chance@lemmy.world 6 months ago
You asked for a line, they gave you a line, what more can be asked for?
We used to intentionally add wrong data to our datasets so we could circle it afterwards, declare it “outlying data” and gain the extra points for spotting and documenting it.
I’m not bitter about my formal education, honest…
Engywuck@lemm.ee 6 months ago
We used to intentionally add wrong data to our datasets so we could circle it afterwards, declare it “outlying data” and gain the extra points for spotting and documenting it.
Nice trick!
rustydrd@sh.itjust.works 6 months ago
If you’re having proportion of explained variance problems I feel bad for you son
I got ninety nine problems but a fit ain’t onemindbleach@sh.itjust.works 6 months ago
Gaussian: “Squint.”
Gork@lemm.ee 6 months ago
Dat spread tho
MajorHavoc@programming.dev 6 months ago
Looks like a successfullybtrained learning model, to me. (Sarcasm)
Flocklesscrow@lemm.ee 6 months ago
I don’t know why you get credit at all; I do all the work. - Excel
collapse_already@lemmy.ml 6 months ago
They’ll also try to linear fit even the most obvious exponential curve.
yesman@lemmy.world 6 months ago
And if that doesn’t work, there’s always factor analysis.
troyunrau@lemmy.ca 6 months ago
How can you argue with a word like "best’ anyway ;)
uis@lemm.ee 6 months ago
Probably minimal surface ellipse.
Bougie_Birdie@lemmy.blahaj.zone 6 months ago
It’s the line of best fit, not the line of good fit
macarthur_park@lemmy.world 6 months ago
Line of “least bad” fit
merari42@lemmy.world 6 months ago
Best Linear Unbiased Estimator