Evaporation and selection effects
A short note on evaporation, selection effects, and why separating molecules dominates the energy accounting.
Toban Wiebe
AI engineering, matching markets, Bayesian thinking, and keyboard-first computing.
Latest posts
Use eCDFs instead of histograms
Histograms are familiar, but empirical CDFs are usually a better default for visualizing distributions: no bin widths, easy percentiles, and better behavior in the tails.
New essay: Use eCDFs instead of histograms
Histograms make you fiddle with bin widths, hide percentiles, and handle tails badly. eCDFs avoid all three problems.
Stop worrying about class imbalance
Data scientists worry too much about class imbalance in classifier training sets. If you care about calibrated probabilities, log loss already handles the imbalance correctly.
New mini-app for name analytics: What's in a Name?
I made a fun data app that guesses whether a name feels old, new, rising, or fading.
New essay: Stop worrying about class imbalance
Class imbalance is mostly a non-issue.
Quantifying uncertainty in probability predictions
Suppose you're interested in knowing the chances of an event X occuring (e.g., X = " a nuclear strike over any populated area in the year 2019 "). When making predictions a...
Quantifying uncertainty in probability predictions
Suppose you're interested in knowing the chances of an event X occuring (e.g., X = " a nuclear strike over any populated area in the year 2019 "). When making predictions a...
Find Your Community: Increasing user engagement at Reddit
Toban Wiebe is an Insight Data Science Fellow in Silicon Valley. For his fellowship project, he performed an analysis of Reddit user engagement to provide actionable insigh...
Dealing with Anaconda Python in Linux
Anaconda is a very handy Python distribution that bundles a slew of scientific packages along with the handy conda package manager, allowing you to easily update your packa...
Make your own Julia packages
Update : with the release of Julia 1.0, the package manager, Pkg , was completely rewritten, though it remains quite similar. Check out the Pkg documentation to see what's...
MathJax with Kramdown
Last week, GitHub Pages upgraded to Jekyll 3.0. One major consequence of the upgrade is that, for Jekyll sites hosted on GitHub Pages, only the kramdown engine is supported...
Simple keyboard shortcuts for any website
I'm a huge fan of Gmail's vim-inspired keyboard shortcuts. It's been nice to see the adoption of keyboard shortcuts in other sites, but for most of the web, keyboard access...
Economic growth via effective regulation
I somehow ended up reading this long essay by economist John Cochrane on growth-oriented policy. I really enjoyed it because it takes a hard-headed approach to topics where...
Economic growth via effective regulation
I somehow ended up reading this long essay by economist John Cochrane on growth-oriented policy. I really enjoyed it because it takes a hard-headed approach to topics where...
Note taking like a hacker
In the spirit of blogging like a hacker, I am now taking notes like a hacker, thanks to a system I pieced together using Markdown and Dropbox. I have used Evernote and Goog...
New Jekyll site
So I decided to move from Wordpress to Jekyll. Not that I had any complaints with Wordpress, it's an absolutely fantastic platform. I just wanted to start blogging like a h...
Double standard: caffeine vs nicotine
[...] there is a kind of puritanical view that everything relating to nicotine is bad and harmful and should be stamped on. - Richard West I'm no expert on this topic, but...
Double standard: caffeine vs nicotine
[...] there is a kind of puritanical view that everything relating to nicotine is bad and harmful and should be stamped on. - Richard West I'm no expert on this topic, but...
A thought on methodology
I've never felt comfortable with the logical positivists' "science is prediction" characterization, primarily because it neglects what I intuitively think of as the heart o...
A thought on methodology
I've never felt comfortable with the logical positivists' "science is prediction" characterization, primarily because it neglects what I intuitively think of as the heart o...