Better Visualizations, Advanced ETL Techniques, RAG Pain Points, and Other February Must-Reads

Author:Murphy  |  View: 20821  |  Time: 2025-03-22 22:41:49

February might be the shortest month, but it certainly didn't feel this way here at TDS, where our authors have been on top of their game, sharing strong contributions on timely topics – including some of the longest and most-read articles of the year so far.

Now that most of us have settled into the flow of things in 2024, we see our readers focus slightly less on career moves and more on core skills and concrete solutions to common issues. Our most-read and -discussed articles of the past month reflect that, and below you'll find a representative sample of our February standouts.

Monthly Highlights

  • The Math Behind the Adam OptimizerIn a clear, accessible, and widely shared explainer, Cristian Leo unpacks the mathematical inner workings of the Adam (Adaptive Moment Estimation) optimizer and, along the way, helps us understand why it's become such a popular choice among deep learning practitioners.
  • 12 RAG Pain Points and Proposed SolutionsWhile retrieval-augmented generation continues to make waves as a powerful option for boosting LLMs' performance, its shortcomings are becoming clearer, too. Wenqi Glantz offers a useful resource for anyone who's felt stuck implementing a RAG system recently, compiling 12 common pitfalls as well as suggested workarounds.
  • Data Visualization 101: Playbook for Attention-Grabbing VisualsFor anyone looking to create "clearer, sharper and smarter visuals"—and who isn't, really?—the latest data-visualization guide by Mariya Mansurova is essential reading, as it leverages numerous concrete examples (in Plotly) to showcase essential design principles in action.

Comment

Recommend

  • ◦ Large Language Models, ALBERT – A Lite BERT for Self-supervised Learning
  • ◦ Getting Started Predicting Time Series Data with Facebook Prophet
  • ◦ Getting Started with Multimodal AI, One-Hot Encoding, and Other Beginner-Friendly Guides
  • ◦ Multinomial Naive Bayes Classifier
  • ◦ The Intuition behind Concordance Index – Survival Analysis
  • ◦ Utilising pykrige and matplotlib for Spatial Visualisation of Geological Variations
  • ◦ Guided Transfer Learning: How to use ‘the power of scouts' to boost machine learning per
  • ◦ SHAP vs. ALE for Feature Interactions: Understanding Conflicting Results
  • ◦ Make Python Faster by Caching Functions: Memoization
  • ◦ Implement Multi-GPU Training on a single GPU
  • ◦ Economics of Generative AI
  • ◦ Creating an Assistant with OpenAI Assistant API and Streamlit