Thanks for visiting my github page. My repos focus on machine learning, statistics, functional programming, puzzle solving, and notes to myself. I tend to program in either python or R. Some repos are full-fledged data science tools and are published outside of github. Others are pet projects.

• python: statsmodels provides classes and functions for the estimation of many different statistical models.
• R: dann is an implementation of Hastie and Tibshirani’s Discriminant Adaptive Nearest Neighbor Classification.
• R: tidydann adds the 'dann' model and the 'sub_dann' model to the Tidymodels ecosystem.
• python: microsoftLTR trains a M.L. model that directly optimizes gain.
• R: survivoR builds time to event models.
• python: anomaly_detection trains multiple anomaly detection models on a simulated dataset.
• R: extendedFamily adds new links to R’s generalized linear models.
• python: translator translates English to Spanish with tensorflow.
• python: glm_irls is an implementation of generalized linear models from the ground up using numpy.
• python: coord-descent-glm is an implementation of generalized linear models using coordinate descent and functional programming.
• python: tensorflow contains examples of
  • two versions of a generative adversarial network
  • transfer learning
  • data augmentation
  • functional A.P.I. with a residual connection
  • auto encoder
• python: aws_docker_py is a containerized model in AWS.
• python: semi_supervised_two explores the usefulness of semi-supervised machine learning.
• python: feature_selection compares different feature selection methods for machine learning.

• R: LRTesteR is a collection of hypothesis tests and confidence intervals based on the likelihood ratio.

• R: TypeOneTypeTwoSim is a simulation of type I error rates, type II error rates, and coverage rates of functions in LRTesteR.
• R: geometric_likelihood_ratio explores a distribution where asymptotic theory does not apply.
• R: calibration studys calibration of p values from likelihood ratio tests when sample size is small.
• R: normalTestsCompare compares power of Gaussian goodness of fit tests.
• R: muTestsCompare compares nonparametric tests for mu.
• R: medianTestsCompare compares nonparametric tests for the median.
• R: bayesian_p_values studies how changing the prior distribution's parameters affects p value calculations.

• R: GlmSimulatoR allows the user to easily and quickly create data for the generalized linear model.
• python: datasets-friedman-1994 implements simulated dataset algorithms from Friedman (1994).

• R: functional_playground contains odds and ends of functional programming ideas.
• R: altForm contains alternative formulations of statistical functions.

• python: backtracking solving puzzles using backtracking algorithms.
  • Sudoku puzzles
  • Knights tour problem
  • N queens problem
  • Pizza Hut's pi day challenge.

• R: glm_notes is a collection of notes about generalized linear models.
• python: interviewQuestions is a collection of technical programming questions I have been asked during data science interviews.
• python: conda_environments contains conda commands for my typical conda environments.
• pencil: proofs is a collection of math proofs.