Entropy calculator
Paste nonnegative counts, weights or probabilities. The calculator normalizes them to sum to one.
Category contributions
| Category | Weight | p | −p log(p) |
|---|---|---|---|
| Fill in the fields above — the result is computed on this page. | |||
How do I use this page?
- Paste weights for all categories.
- Choose the logarithm base.
- Read H, normalized entropy and each category contribution.
What this page does not do: Entropy describes category shares but ignores ordering and similarity between categories. Small-sample estimates can be biased.
Further reading
Theory — what does Shannon entropy measure?
Shannon entropy measures the uncertainty or diversity of a data distribution. It equals zero when all the mass falls on a single category and reaches its maximum when every share is equal. With a logarithm base of 2 the result is expressed in bits, and normalized entropy divides that value by the maximum possible for the given number of categories. You can paste either raw counts or ready-made probabilities into the calculator — it normalizes them to sum to one automatically.
Frequently asked questions
Can I enter counts instead of probabilities?
When is entropy largest?
See also
A/B test calculator
Compare two proportions in an A/B test. Calculate conversion rates, difference, uplift, z score, p value and c
A/B test planning — sample size, MDE and duration
Plan a two-proportion experiment, estimate detectable effect and traffic duration.
ACF and PACF autocorrelation calculator
Calculate ACF and PACF across time-series lags. Review the chart, reference bounds and coefficient table.