Start here in Data Science
The most useful guides in this topic, not just the most recent.
51 Free Guides
Data Science Guides & Tutorials
Roadmaps, salaries, projects and interview prep for aspiring data scientists — from statistics and machine learning to model deployment.
Data Science articles
Hyperparameter Tuning: Grid Search vs Random Search vs Bayesian
Grid search, random search and Bayesian optimisation compared on one real model with the same budget of 24 candidates: how close they came, why the differences sat inside the noise between folds, why random search explores more, and a second model where tuning made the difference between 61% and 93% accuracy.
Top 10 Machine Learning Algorithms Explained for Beginners
The ten machine learning algorithms beginners should know, each explained by how it works, what it is good at and where it fails, plus a head-to-head benchmark of the seven classifiers on three real datasets under identical cross-validation, and what published studies of hundreds of datasets add to that picture.
Time Series Forecasting: A Beginner-Friendly Introduction
A beginner-friendly introduction to time series forecasting on 36 years of real monthly CO2 measurements: what trend and seasonality are, why simple benchmark methods matter, how Holt-Winters combines both patterns, and how to test forecasts honestly by repeatedly forecasting the next year from past data only.
How to Build a Data Science Portfolio That Gets Interviews
How to build a data science portfolio designed for the few minutes a reviewer spends on it: which three projects to show, a README template that puts the result first, repository habits that signal professional work, when a live demo is worth it, and the red flags that get portfolios skipped.
How to Read a Machine Learning Research Paper
How to read a machine learning research paper without drowning in it: the three-pass method applied to the Transformer paper, the questions to ask of every results table, the difference between what a paper shows and what it suspects, and a small NumPy re-implementation that checks one of its claims.
Computer Vision Basics: From Pixels to Predictions
Computer vision from the ground up: what an image is to a computer, how a convolution filter finds edges, a small convolutional network trained on handwritten digits against a pixel baseline, the main vision tasks from classification to segmentation, and why transfer learning is the default.