By no means miss a brand new version of The Variable, our weekly publication that includes a top-notch collection of editors’ picks, deep dives, group information, and extra.
With the top of the yr only a few weeks away, neither our authors nor our readers are displaying any indicators of slowing down.
We’re thrilled to have revealed a few of our strongest articles of the yr previously month: sensible guides on LLM workflows and sources on profession progress, Python-focused tutorials, and deep dives on just lately launched instruments, amongst different standout matters. Learn on to meet up with (or revisit) November’s most-read tales.
Graph RAG vs SQL RAG
Which database paradigm delivers extra correct and insightful outcomes? Reinhard Sellmair units out to evaluate the efficiency of two sorts of RAG techniques by pitting GraphRAG and SQL RAG towards one another, utilizing the identical dataset and questions.
LLM-Powered Time-Collection Evaluation
Within the second a part of Sara Nobrega’s common collection, we be taught concerning the prompts we want for superior mannequin improvement (assume ARIMA and LSTM).
The right way to Construct Machine Studying Tasks That Assist You Get Employed
Not all ML portfolios are created equal. Egor Howell shares time-tested insights on what works — and what doesn’t.
Different November Highlights
Don’t miss our different prime reads from the previous month, tackling NumPy, Multimodal RAG, marimo notebooks, and plenty of different matters — each evergreen and leading edge.
NumPy for Absolute Learners: A Venture-Based mostly Strategy to Knowledge Evaluation, by Ibrahim Salami
Understanding Convolutional Neural Networks (CNNs) By Excel, by Angela Shi
Run Python As much as 150× Sooner with C, by Thomas Reid
The right way to Construct an Over-Engineered Retrieval System, by Ida Silfverskiöld
Constructing a Multimodal RAG That Responds with Textual content, Pictures, and Tables from Sources, by Partha Sarkar
Why I’m Making the Swap to marimo Notebooks, by Parul Pandey
Your Subsequent ‘Massive’ Language Mannequin Would possibly Not Be Massive After All, by Moulik Gupta
In Case You Missed It: Our Newest Creator Q&As
We love sharing our authors’ experience, profession insights, and views on the latest developments on the planet of knowledge science and AI. Listed here are our most up-to-date Creator Spotlights.
- “Techniques pondering helps me put the massive image entrance and heart”
Shuai Guo on deep analysis brokers, analytical AI vs LLM-based brokers, and techniques pondering.
- “The success of an AI product relies on how intuitively customers can work together with its capabilities”
Janna Lipenkova on AI technique, AI merchandise, and the way area data can change the complete form of an AI answer.
Meet Our New Authors
We hope you’re taking the time to discover the superb work from the newest cohort of TDS contributors:
- Jure Leskovec, a Stanford professor of pc science and entrepreneur, explains why LLMs aren’t a one-size-fits-all answer for corporations.
- Sherin Sunny, a senior engineer at Walmart, walked us by means of the creation of a pc imaginative and prescient undertaking geared toward detecting leaves.
- Manuel Franco de la Peña launched us to ShaTS, a novel Shapley-based explainability methodology particularly designed for time-series fashions, which he co-created.
We love publishing articles from new authors, so in case you’ve just lately written an attention-grabbing undertaking walkthrough, tutorial, or theoretical reflection on any of our core matters, why not share it with us?
We’d Love Your Suggestions, Authors!
Are you an current TDS writer? We invite you to fill out a 5-minute survey so we will enhance the publishing course of for all contributors.
