Home › Key concepts › Start with a baseline

Key concept · Crawl (beginner) · the same on every platform

Start with a baseline

Beat a simple rule first, and say by how much.

What it is

A baseline is the simplest reasonable predictor, such as 'predict last month's value' or 'predict the most common class'. Every model is judged by how much it improves on it.

Why it matters

It shows whether machine learning is needed at all, and stops impressive-sounding scores that are no better than guessing.

An example

95% accuracy sounds great until the baseline 'nobody churns' scores 94%.

The common mistake

Reporting accuracy with no baseline beside it.

Where you meet it on each platform

The idea stays the same; only the product name changes.

JobAWSAzureGoogle CloudDatabricksOpen source
Trainingfit a model on data, on a laptop or a clusterSageMaker training jobsAzure ML jobsVertex AI trainingDatabricks ML runtimescikit-learn, PyTorch

Related ideas

Where it shows up

Part of the Crawl level on these paths: ML Engineering (learning basics).