An improved git flow, probabilistic programming, better status updates
-
We use the common git flow described in the article below, and we suffer from precisely the drawbacks discussed (multiple commits per MR/PR, tedious code reviews). Annie Sexton‘s Better Git Flow recommends adding a simple
git reset
step to clean things up. I love it. -
A walkthrough of how Nathan Cooper Jones tuned a GAN architecture to generate 100,000 Bored Apes that do not exist.
-
If, like me, you don’t come from a stats or Bayesian background, but still find yourself working with probabilistic models, it can be troubling not to know what’s going on in the code. Working through this example of a probabilistic programming language in 70 lines of Python was instructive and helps to reinforce the concepts that I find myself reviewing periodically.
-
Streamlit’s blog has a very simple walkthrough of creating a streamlit app.
-
Improve your weekly (or daily) status updates: focus on impact.
-
Jakob Greenfield posted a pretty simple exercise on how he learned to love his monkey mind.
And related, One Question from James Clear worth writing on a post-it note: How can I make the next 60 seconds as perfect and as present as possible?
And another: A Simple Math Equation That Can Transform Your Productivity
“Most of us go through life operating at a fraction in everything we do.”