Published inTowards Data Science·PinnedMember-onlyExploring GEMBA: A New LLM-Based Metric for Translation Quality AssessmentUsing LLMs for evaluating translation quality — Introduction I recently read an intriguing paper from the Microsoft¹ team (published in May 2023) that caught my attention. The paper delves into the world of translation evaluation, shedding light on an innovative metric called GEMBA (GPT Estimation Metric Based Assessment). In this blog post, we’ll dissect the paper and provide…Language Translation9 min readLanguage Translation9 min read
Published inTowards Data Science·PinnedA Gentle Intro to Chaining LLMs, Agents, and utils via LangChainUnderstand the basics of agents, tools, and prompts and some learnings along the way — Audience: For those feeling overwhelmed with the giant (yet brilliant) library… Introduction I’d be lying if I said I have got the entire LangChain library covered — in fact, I am far from it. But the buzz surrounding it was enough to shake me out of my writing hiatus and give…Large Language Models20 min readLarge Language Models20 min read
Published inTowards Data Science·PinnedMember-onlyHow to Make the Most of Your Python Debugger in VSCodeWatch variables, use conditional breakpoints, call stack, exception breakpoint, etc— when working on larger projects — Quick pulse check — Still using print() statements to debug your code? Hey, no shame in admitting if you do. If it works, it works (especially for smaller scripts)! Having said that, it’s not a sustainable solution when working with larger projects — involving a buttload of third-party packages, util…Vscode13 min readVscode13 min read
Published inTowards Data Science·PinnedMember-onlyData scientist’s guide to efficient coding in PythonTips and tricks I use for writing clean codes on a day-to-day basis — In this article, I wanted to share a few tips for writing cleaner codes that I have absorbed in the last year — mainly from pair programming. …Python11 min readPython11 min read
Published inTowards Data Science·PinnedMember-onlyTime Series Analysis using Pandas in PythonBonus intro to keywords like seasonality, trend, autocorrelation, and much more. — Right off the bat, time-series data is not your average dataset! You might have worked with housing data wherein each row represents features of a particular house (such as total area, number of bedrooms, year in which it was built) or student dataset wherein each row represents such information about…Time Series Analysis14 min readTime Series Analysis14 min read
Published inThe Startup·Sep 133 Apps to Get You Smarter FasterAnd no I am not telling you to go read a book — In a world where continuous learning is not just a choice, but a necessity, the quest for getting smarter has never been more exciting. It is also an important means for escaping the infamous imposter syndrome which has now become the standard rather than an exception. …Productivity5 min readProductivity5 min read
Jul 615 Best History Podcasts of 2023History of the world — recent and ancient. — This week, I made a conscious decision to take a break from my usual routine of writing technical blogs. Instead, I opted for a rejuvenating hobby that allows me to unwind and de-stress after a long day: listening to captivating podcasts. Engaging with these podcasts has become a form of…Podcast7 min readPodcast7 min read
Published inTrusted Data Science @ Haleon·May 30Deriving insights from customer queries on Haleon brandsHow to build a dashboard for generating insights — by Dr. Varshita Sher, Zohaib Khan Introduction Haleon has a world-class portfolio of category-leading global brands, including Sensodyne, Voltaren, Panadol, and Centrum, to name a few. Our brands are trusted by health professionals, customers, and people worldwide to improve the health and well-being of individuals and their communities. It is, therefore…NLP5 min readNLP5 min read
Nov 30, 2022Using Causal ML Instead of A/B Testing1.1K17Samuele MazzantiThanks, Samuele.Thanks, Samuele. This is brilliant. Clearly the cut off assumption plays a big role here in generating the predictor matrix. Do you have any recommendations in this regard - perhaps trying a bunch of cut offs, averaging, etc and the impact on the outcome, if any?1 min read1 min read
Nov 11, 2022Why use `mono=False` when loading the audio with librosa?1Cleo LafayetteThat's cause we don't want to average samples across different channels (and risk losing some…That's cause we don't want to average samples across different channels (and risk losing some signal in data) and ultimately convert the audio into 1-D array (especially when HuBERT can handle multi-dimensional audio arrays).1 min read1 min read
