TIL about FiraGO. Fira is certainly my second preferred Sans (and Code) font after Iosevka.

R packages can be just a way to organise code, nothing more. By following a certain structure and a few rules we get to benefit from a whole bunch of tools designed just for packages. — Data Science Workflows.

On Git.

• Git: Best Practices
• Git Best Practices and Tips by Toptal Developers
• Git commit practices your future self will thank you for
• Commit Often, Perfect Later, Publish Once: Git Best Practices

Draw Lindenmayer Systems with Common LISP. #lisp

The Y Combinator Explained in Python.

  Nick Cave & The bad Seeds, The Abattoir Blues Tour.

Why linear mixed-effects models are probably not the solution to your missing data problems. #rstats

On Moving from Statistics to Machine Learning, the Final Stage of Grief.

Slowly going back to my unread feeds in Reeder.app. Today is After WWDC 2020: bittersweet Mac. #apple

Logistic regression is not fundamentally a classification algorithm.

The problem arises from logistic regression often being taught as a “classification” algorithm in the machine learning world. I was personally not taught this way– I learned from econometricians that you can use either probit or “logit” as general linear models in the event you want to estimate on a binary target variable, and that these models calculate probabilities. Thinking about logistic regression as a probability model easily translates to the classification case, but the reverse simply does not seem to be true. — Why Do So Many Practicing Data Scientists Not Understand Logistic Regression?