Deploying MLFlow
Remember kids, the only difference between screwing around and science is writing it down. Adam Savage
Recent Posts
Creating Probabilistic Pandas Datatypes
Imagine you’re doing some kind of tabluar data work in Pandas, but rather than work with the same old concrete types, you’re working with some kind of uncert...
Parallel Computing on Windows with R
By default, R doesn’t acknowledge that your computer has multiple logical or even physical cores, which will often result in your long repetitive job tying u...
Tillé’s Method for Unequal Probability Sampling Without Replacement
In sampling from a population, oftentimes one will find that different subclasses of population members occur with different probability, and so it’s useful ...
Shared Keys For File Syncing
I do my work on a few different machines at once, and in order to keep things reasonably sane I like to sync my client machines back to my share hosted on th...