TTharunintharunpandya.hashnode.dev·16h ago · 10 min readPolynomial Regression: Teaching a Line to BendTake our best straight-line model so far, the one that fit our five-order table perfectly. Now widen the range a bit. Some orders close by, a couple kilometers away. Some far, ten or twelve kilometers00
NANethmi Amashaincuriousdeveloperbyamasha.hashnode.dev·Aug 3 · 4 min readWhy Do Both Data Preprocessing and Feature Engineering Handle Missing Values?While studying the Machine Learning Pipeline, I came across two stages that confused me: Data Preprocessing Feature Engineering At first, everything seemed clear. Data Preprocessing is responsible00
SGSergio González Téllezinevankhandev.hashnode.dev·Jul 12 · 6 min readBeyond SQL: How a Data Science Capstone Changed the Way I Think About Metrics, State, and Observability1. Introduction When I enrolled in the Learn SQL Basics for Data Science Specialization from the University of California, Davis, my goal was straightforward: learn SQL and strengthen my data analysis00
AJakshat jaininakshat-ml.hashnode.dev·Jun 25 · 6 min readFrom 1.4M NYC Taxi Trips to Production ML: Feature Engineering Improved Predictions by 34.6%How thoughtful feature engineering and SHAP explainability turned raw taxi data into a deployed prediction model. The Problem Nobody Thinks About Every time you request a ride in New York City, an alg20
YPYejun Parkinyejunpark3.hashnode.dev·May 30 · 13 min readFrom Thousands of Features to a Few: A Deep Dive into Dimensionality ReductionImagine you're trying to describe a person to a friend. You could list ten thousand physical measurements—exact hair strand thickness, each fingernail length, the curvature of every knuckle—or you cou00
Ddanieljasmeindanieljasme.hashnode.dev·May 27 · 9 min readBuilding a Near-Perfect Claims Classifier: Random Forest vs XGBoost on TikTok Data (Google Advanced DA Course 5)This is the fourth and final post in my Google Advanced Data Analytics Certificate series on the TikTok claims classification project. Part 1: data inspection. Part 2: EDA and visualization. Part 3: h00
SMScott McMahaninaitransformeronline.hashnode.dev·May 4 · 2 min readWhy AI Projects Struggle in ProductionA lot of AI systems look great during development but fall apart once they hit production. The issue is rarely the model itself. It usually comes down to inconsistencies in how data is handled across 00
KEkingsley ebube onohinkingsleyonoh.hashnode.dev·Apr 18 · 6 min readWhy I Split Minutes Prediction Into Two Models Instead of OneA single regression model trained on NBA game logs predicts that Joel Embiid will play 11 minutes in a game where he's listed as OUT. The model has never seen a confident zero. Every row in the training data has some minutes played, because the stand...00
HHHenry Haindatabeginnersfriendlytutorials.hashnode.dev·Apr 15 · 26 min readCross-Validation, Pipelines & Hyperparameter Tuning with Scikit-learnWhen building a machine learning model, it's tempting to train it on your data, check the score, and call it done. But that score is misleading — the model has already "seen" the data it's being evalu00
PGPrajakta Gunjalinprajakta-aiml.hashnode.dev·Apr 12 · 14 min readHow I Built an ML Model to Classify Pancreatic Cancer Subtypes from 60,000 GenesWhere and how did it all start Imagine being handed a spreadsheet with 60,000 columns and being told — somewhere in here is the answer to classifying a cancer that kills 9 out of 10 patients within fi00
