posts – R'tichoke

Tabular Foundation Models: A First Look with TabICL

R

Machine Learning

Foundation Models

reticulate

What tabular foundation models actually do, why in-context learning is a real shift for tabular ML, and how TabICL compares to XGBoost on a credit-risk dataset — without a single hyperparameter tuning loop.

Machine Learning Frameworks in R

R

Machine Learning

R’s ecosystem offers a rich selection of machine learning frameworks, each with distinct design philosophies and strengths. This post is a side-by-side comparison of five ML…

Text Analytics in R: Dense embeddings and RAG pipeline with `ragnar` and `ellmer`

R

NLP

RAG

LLMs

ragnar

ellmer

This earlier post explored building a text analytics pipeline using quanteda. We created a Document Feature Matrix (DFM), weighted it with TF-IDF, and measured document…

Text Analytics in R: Using quanteda

R

Text Analytics

quanteda

NLP

In this post, let’s explore how to transform text data into meaningful insights using the quanteda package — a powerful and intuitive framework for text analysis.

Explore Neural Networks Interactively with Quarto Live!

R

Quarto

Neural Networks

WebR

Visualization

Quarto Live combines Quarto with WebR to enable interactive R code execution directly in the browser allowing for:

Multi-Task Learning with torch in R

R

Deep Learning

torch

Multi-Task Learning

Multi-task learning (MTL) is an approach where a single neural network model is trained to perform multiple related tasks simultaneously. This methodology can improve model…

Building a Simple Neural Network in R with torch

R

Deep Learning

torch

The torch package brings PyTorch to R. In this post, let’s build and train a simple neural network from scratch.

Assessing Credit Score Prediction Reliability Using Bootstrap Resampling

R

Credit Risk Analytics

Bootstrapping

Credit scoring models tend to perform well in the middle of the score distribution — but reliability drops at the extremes, where data thins out. In this post, let’s use…

Building Models in R with tidymodels

R

Machine Learning

tidymodels

The tidymodels framework provides a cohesive set of packages for modeling and machine learning in R, following tidyverse principles. In this post, let’s build a realistic…

Optimizing XGBoost Hyperparameters Using Bayesian Optimization in R

R

Analytics

Machine Learning

Hyperparameter tuning is computationally intensive and time-consuming. In this post, we’ll use Bayesian optimization to search for good XGBoost hyperparameters — a…

Implementing Particle Swarm Optimization from Scratch in R

R

Optimization

Visualization

Nature-inspired optimization algorithms can be surprisingly effective at navigating complex, high-dimensional search spaces. In this post, let’s build a Particle Swarm…

Developing Custom Charting Functions with ggplot2

R

Data Visualization

ggplot2

R has plenty of visualization options, but ggplot2 stands out for its flexibility, quality, and composability. In this post, let’s build reusable custom charting functions…

Generating Correlated Random Numbers in R Using Matrix Methods

R

Statistics

Simulation

Generating random data with a specific correlation structure is a common need in statistical simulation. In this post, let’s walk through how to do it in R using matrix…

Evaluating Binary Classification Models Using Gains Tables

R

Credit Risk Analytics

Model Evaluation

Gains tables (also known as KS tables) are a go-to tool for evaluating binary classification models in credit risk. In this post, let’s build one from scratch in R and…

Portfolio Optimization Using PSO

R

Finance

Optimization

Portfolio optimization represents a critical task in investment management, where the goal involves allocating capital across different assets to maximize returns while…

Introduction to R for Analytics

R

Analytics

Introduction

R is a powerful language built for data analysis and visualization. This post walks through practical examples of using R for real-world analytics tasks.

Monotonic Binning Using XGBoost

R

Credit Risk Analytics

XGBoost

This post focuses on how to implement monotonic binning, a method that groups variable values into bins where event rates demonstrate consistent monotonic behavior. This…

Getting Started with Python using R and Reticulate

R

Python

reticulate

Want to use Python’s powerful libraries without leaving R? The reticulate package gives you the best of both worlds - R’s elegant data handling and visualization with…