🧵 on Linear Regression and R-squared with #rstats & #tidyverse. You have data, but need a model to either predict new points or explain the relationship. Several lines, sales=m(temp) + b, seem reasonable. Which is best?

Remembering our former coach Bob Schul - let's take a look at his big kick to win Olympic gold on a wet track in Tokyo in 1964 - the only American to have won an Olympic gold medal in the 5000m! Here's the final 3 laps - https://t.co/Q1lIFeZNam

lol who did this

I swear the use of "inference" in machine learning was designed to make statisticians as confused as possible

Powerful visuals.

why no-code never took off, in two sentences.

September is childhood cancer awareness month. It's a great time to sign up for the Nov 11th Ezra J Hartke Race for Hope, raising money for pediatric brain tumor research. @DaytonChildrens #daytonoh #childhoodcancerawareness https://t.co/KxP8Ig4GdZ

I’m worried that the push to apply LLMs everywhere will steer young ML folks away from our true calling (whining about data quality)

this is a must watch for every body in the trenches today: - "Do you view this season as a failure?" giannis: "Do you get a promotion every year?" - "No." g: "So every year you work is a failure? " g: "It's not failure - it's steps to success." https://t.co/YHkcFOiRbQ

new website just dropped @daniela_witten @HastieTrevor @robtibshirani and welcoming Jonathan Taylor to the team! https://t.co/0NfCas72Uy

Lot of wisdom in this joke.

AI Risk Management Framework from the US National Institute of Standards & Technology (NIST). Helpful if you or your company is wondering how to define AI/ML systems and manage risk/governance. https://t.co/e45qXUBvPj

Prediction intervals for machine learning models. MAPIE works with any scikit-learn model & is a big improvement over basic methods that only gave constant interval widths. Based on conformal prediction. Awesome library. https://t.co/NGZJaA79I6

Prediction intervals for machine learning models. MAPIE works with any scikit-learn model & is a big improvement over basic methods that only gave constant interval widths. Based on conformal prediction. Awesome library. https://t.co/NGZJaA7HxE

Yes!

Readers, if you received a copy of Data Head with copy-editing errors, DM me.

Any insight from fellow authors? @joshuastarmer, @storywithdata, @fchollet, @aureliengeron, @__mharrison__

Frustrating for readers and authors: poor quality copies on the market. Are these knock-offs or random problems at the publisher? Have reached out to Wiley.

For 2023, stop taking online courses and start reading these wonderful books. #datascience #ai #deeplearbing @__mharrison__ @rasbt @StefanieMolin @BecomingDataSci @GaborBekes @GutmanDataHead @fchollet @aureliengeron @joshuastarmer @billfranksga

Always appreciate when @KirkDBorne shares his recommendation for "Data Head."

Makes me think of arguing with conspiracy theorists. But I still do it, just in case some have a prior > 0.