By-hand code for models and algorithms. An update to the 'Miscellaneous-R-Code' repo.

Animated Visualizations of Popular Machine Learning Algorithms

Gaussian mixture modelling - Unsupervised learning

Designing and applying unsupervised learning on the Radar signals to perform clustering using K-means and Expectation maximization for Gausian mixture models to study ionosphere structure. Both the algorithms have been implemented without the use of any built-in packages. The Dataset can be found here: https://archive.ics.uci.edu/ml/datasets/ion…

R, Julia and Python implementation of the two submarket fully endogenized finite mixture model used in forthcoming articles by Fuad and Farmer (202-) and Fuad, Farmer, and Abidemi (202-).

The multi-sample Gaussian mixture model (MSGMM) is a clustering model adapted to fitting multiple samples simultaneously using the EM algorithm.

Code for our ICML '18 paper "Closed-form Marginal Likelihood in Gamma-Poisson Matrix Factorization"

🌾 EDA and several Clustering techniques applied on a real dataset of wheat kernels. 🍞

Gibbs Sampling & EM Algorithm Implementation / R Programming Language / on Iris Dataset

BEEM: Estimating Lotka-Volterra models from time-course microbiome sequencing data

K-means and EM from scratch. A short discussion of their differences and performance.

From-scratch implementation of Multivariate Expectation-Maximization algorithms.

A nucleosomal reads counting tool

R code for Panel Data Analysis with Expectation Maximization iteration for missing values. Final models are heteroskedastic, so Robust Covariance Matrix is used. Since the panel analysis didn't produce any potential fixed effect models, Chow Test wasn't included.

Expectation-Maximization for Semi-Supervised Learning, where only a sample of observations are categorized. Code is written in R and reproduced in Python. Documentation discusses the theoretical background of the algorithm, as well as applications in Fisheries and Insurance Pricing.

Data Clustering using Expectation Maximization algorithm. To cite this Original Software Publication: https://www.sciencedirect.com/science/article/pii/S2352711021001771

R package for fitting mixture distributions to data using various approaches