Replay a recorded greenhouse sample

A journey through can we repeat the same experiment? with Botie.

Python and NumPy · A journey with Botie

First visit? Learn the workshop controls

Botie's notebook

Giving Simulated Experiments a Repeatable Starting Point

Part 1 of 2

Numerical simulations need many values that behave like random draws, yet debugging also requires the ability to repeat an experiment. Pseudorandom generators address this by maintaining a state and updating it through a deterministic rule.

Sources for this history

NumPy random sampling
PCG64 and reproducible streams

← Previous lesson: When Two Calculations Almost Agree Read the full lesson and derivations