liveBook · Manning

Odds

As explained in Grokking Machine Learning

Odds are the ratio of the probability of an event occurring to the probability of it not occurring.

Odds compare how often an event happens with how often it does not. If its probability is p, its odds are p/(1 − p), so odds range from zero to infinity. section 12.3

A die and a horse race

Rolling a 1 on a fair die has probability 1/6 and odds of 1/5. A horse that wins three out of four races has probability 3/4 and odds of 3 to 1. section 12.3

In AdaBoost, odds lead into log-odds, or the logit, which scores weak learners according to their accuracy. section 12.3

How Grokking Machine Learning covers it

How deep it goes
Defines odds with a formula, gives two probability examples, and notes their range before moving to log-odds.
Examples
Illustrates odds with rolling a 1 on a die and a horse winning three of four races.
What it assumes
Assumes familiarity with probability and explains odds through event and non-event counts.
What you can do after
Evaluate an event’s odds from its probability and distinguish odds from probability.
How it connects
Connects odds to probability, log-odds, the logit, and AdaBoost weak learners.
Cover of Grokking Machine Learning
liveBookmanning.com
Read Grokking Machine Learning on liveBook Get the book at manning.com

Also explained in

Compare how 3 books explain odds →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Grokking Machine Learning. The section links point to the parts of the book it draws on.