Implementation:Online ml River Utils Random

Knowledge Sources	Online_ml_River
Domains	Online_Learning, Random_Sampling, Statistics
Last Updated	2026-02-08 16:00 GMT

Overview

Random variable generation for Poisson and exponential distributions.

Description

Provides efficient implementations of poisson and exponential random variate generators. These avoid overflow issues present in naive implementations and are optimized for online learning scenarios where random sampling is needed.

Usage

Use for generating inter-arrival times (exponential) or count data (Poisson) in simulations, synthetic data generation, or stochastic processes. Essential for modeling event streams and time-based phenomena.

Code Reference

Source Location

Repository: Online_ml_River
File: river/utils/random.py

Signature

def poisson(rate: float, rng=random) -> int:
    ...

def exponential(rate: float = 1.0, rng=random) -> float:
    ...

Import

from river import utils

Usage Examples

import random
from river import utils

rng = random.Random(42)

# Poisson samples (count data)
print("Poisson samples (rate=3):")
for _ in range(5):
    sample = utils.random.poisson(rate=3, rng=rng)
    print(sample, end=' ')
print()

# Exponential samples (inter-arrival times)
print("\nExponential samples (rate=0.5):")
for _ in range(5):
    sample = utils.random.exponential(rate=0.5, rng=rng)
    print(f"{sample:.2f}", end=' ')
print()

# Simulating event stream
print("\nEvent arrival times:")
time = 0
for _ in range(10):
    inter_arrival = utils.random.exponential(rate=2.0, rng=rng)
    time += inter_arrival
    n_events = utils.random.poisson(rate=1.5, rng=rng)
    print(f"Time {time:.2f}: {n_events} events")

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment