RAGAs (Retrieval-Augmented Generation Assessment) is a framework (GitHub, Docs) that provides you with the necessary ingredients to help you evaluate your RAG pipeline on a component level. Evaluation Data What’s interesting about RAGAs is that it started out as a framework for “reference-free” evaluation [1]. That means, instead of having to rely on human-annotated ground…

Demystifying the confusion about Linear Regression Visually and Analytically Image: Linear regression illustration, by Stpasha, via Wikimedia Commons (Public Domain). Original Image Link: https://upload.wikimedia.org/wikipedia/commons/8/87/OLS_geometric_interpretation.svgThe above image represents a geometric interpretation of Ordinary Least Squares (OLS) or Linear Regression (words used interchangeably in classical statistics). Let’s break down what we’re seeing in an intuitive way: Variables…

Using reinforcement learning from scratch to teach a computer to play Tic-Tac-Toe Image by author (created with ChatGPT)It appears that everyone in the AI sector is currently honing their Reinforcement Learning (RL) skills, especially in Q-learning, following the recent rumours about OpenAI’s new AI model, Q* and I’m joining in too. However, rather than speculating…

A guide for developers and architects especially those who are not specialized in machine learning to design an MLOps architecture for their organization Introduction According to Gartner’s findings, only 53% of machine learning (ML) projects progress from proof of concept (POC) to production. Often there is a misalignment between the strategic objectives of the company…

The dos and don’ts of processing data in the age of AI 10 min read · 23 hours ago Image by Gilles Lambert on UnsplashThe digital economy has been built on the wonderful promise of equal, fast, and free access to knowledge and information. It has been a long time since then.…

Where the assumptions behind the two-tower model architecture break — and how to go beyond (Image created by the author using generative AI)Two-tower models are among the most common architectural design choices in modern recommender systems — the key idea is to have one tower that learns relevance, and a second, shallow, tower that learns…

A Spotify Data Scientist’s guide to developing a job-hunting strategy that can get you offers A snapshot of the survey I conducted on Blind — Image by AuthorI recently ran a survey among data scientists and found out this shocking number — 86% are blindly sending out job applications, and hoping for the best. Hoping…

Learn about Radial Treemaps and create your own with Python Radial-Treemap by Nick GerendThe Treemap Concept The “Treemap” was introduced by Ben Shneiderman at the University of Maryland in the early 1990s¹. Simply put, it’s an efficient way of displaying hierarchical data as a set of nested rectangles. Although the concept is simple, the arrangement…

Reinforcement learning without the boilerplate code Created by the author with Leonardo Ai.In my previous articles about reinforcement learning, I have shown you how to implement (deep) Q-learning using nothing but a bit of numpy and TensorFlow. While this was an important step towards understanding how these algorithms work under the hood, the code tended…

Three stories about the data career journey “The number 12 is considered a cosmic number — marking the 12 months, the 12 signs of the zodiac, and the 12 stations of the Moon and the Sun” [Photo by Michael Skok on Unsplash]In my previous position, I created end-of-year reports for my business colleagues and CEOs.…