Projects


How I spend my time when I'm not working

Python


gym-inventory

gym-inventory is a single agent domain featuring discrete state and action spaces that an AI agent might encounter in inventory control problems. The vignette motivating this environment is a version of the inventory control with lost sales problem described in Example 1.1 of Algorithms for Reinforcement Learning by Csaba Szepesvari (2010).

gym-gambling

gym-gambling is a single agent domain featuring both discrete and continuous state and action spaces that an AI agent might encounter in gambling problems. The vignette motivating this environment is a version of the gambling problem described in Example 1.2 of Algorithms for Reinforcement Learning by Csaba Szepesvari (2010).

R


scorer

An R utility package for scoring models in data science and machine learning. This toolset is written in C++, where possible, for blazing fast performance. This toolset’s API follows that of Python’s sklearn.metrics as closely as possible so one can easily switch back and forth between R and Python without too much cognitive dissonance.

titanic

An R utility package containing data sets providing information on the fate of passengers on the fatal maiden voyage of the ocean liner “Titanic”, with variables such as economic status (class), sex, age and survival. These data sets are often used as an introduction to machine learning on Kaggle. More details about the competition can be found here, and the original data sets can be found here.

functools

An R utility package for doing Functional Programming in R.

anonymizer

An R utility package for anonymizing data containing Personally Identifiable Information (PII) using a combination of salting and hashing. You can find quality examples of data anonymization in R here, here, and here.

Julia


MLMetrics.jl

A Julia utility package for scoring models in data science and machine learning. This toolset is written in Julia for blazing fast performance. This toolset’s API follows that of Python’s sklearn.metrics as closely as possible so one can easily switch back and forth between R and Python without too much cognitive dissonance.