Skip to content

Numerical Python·Lesson 25 of 67·intermediate

Random number generation and seeding

The modern Generator API, why a global seed is a trap, and how to sample tokens from a distribution with a temperature — reproducibly.


In this lesson · 16 min

  • →Create a seeded Generator with default_rng and draw from it
  • →Explain why a seeded run reproduces and an unseeded one cannot
  • →Keep independent streams independent instead of sharing a global one
  • →Shuffle and split a dataset reproducibly
  • →Sample from a probability distribution, including temperature sampling

Randomness is everywhere in machine learning: the weights you start from, the order you see examples in, the split between train and test, the token a model picks. Controlling it is what separates a result you can defend from one you cannot repeat.

The Generator API

PythonPython not started

The seed fixes the entire stream. Two generators made from the same seed produce identical sequences, for ever:

PythonPython not started

Independent streams

When several parts of a pipeline need randomness, give each its own generator rather than interleaving draws from one. spawn derives children that are statistically independent and still fully determined by the parent seed:

PythonPython not started

Distributions

PythonPython not started

Weighted choice draws from a distribution you supply:

PythonPython not started

Shuffling and splitting a dataset

Shuffle indices, not the data, so several parallel arrays stay aligned:

PythonPython not started

Sampling with a temperature

This is how a language model picks its next token. The logits become probabilities through a softmax, and temperature controls how sharp that distribution is:

PythonPython not started

Low temperature concentrates the mass on the top token and the output becomes nearly deterministic; high temperature flattens it and unlikely tokens start appearing.

PythonPython not started

Greedy decoding is the limit as temperature approaches zero — just argmax, no sampling at all:

PythonPython not started

Masking with -inf works because np.exp(-inf) is exactly 0, so those tokens can never be drawn.

Try it yourself

Exercise 1

Split the ten examples into an 80/20 train and test set using a generator seeded with 7. Shuffle the indices so features and labels stay aligned.

Your turnPython not started
Exercise 2

Write sample(logits, temperature, rng): apply the temperature, turn the logits into probabilities with a stable softmax, and return one index drawn from that distribution.

Your turnPython not started

Check yourself

Why prefer np.random.default_rng(seed) over np.random.seed(seed)?

Check yourself

What does raising the sampling temperature from 0.5 to 2.0 do to the token distribution?

Recap

  • np.random.default_rng(seed) gives a private, seeded generator. Pass it around; do not reach for the global np.random functions.
  • The same seed reproduces the same stream exactly — which is what makes a reported score checkable.
  • parent.spawn(n) derives independent child streams from one seed.
  • Shuffle indices and apply the same order to every parallel array, so rows keep their labels.
  • A temperature divides the logits before the softmax: low sharpens towards greedy, high flattens. Masking with -inf removes a token entirely, because exp(-inf) is 0.