Reference library

A/B testing & experimentation

User bucketing, experiment metrics, statistical comparison, and rollout guardrails.

3 matches

How to simulate a contextual bandit in Python

Simulate a contextual multi-armed bandit with random features and epsilon-greedy action selection in Python.

bandit-algorithms simulation epsilon-greedy

import random


class ContextualBandit:
    def __init__(self, n_actions=3, n_features=4):
        self.n_actions = n_actions
        self.n_features = n_features
        self.theta = [random.random() for _ in range(n_actions * n_features)]

    def mock_context(self):
        return [random.uniform(-1, 1) for _ in ra…

13 0 Open

A/B testing & experimentation medium

Thompson Sampling Mock Bandit in Python

Implement a Thompson sampling multi-armed bandit to explore and exploit reward probabilities across multiple options, updating Beta distributions over time.

thompson-sampling bandit-algorithms exploration-exploitation

Python

import random

class ThompsonSamplingBandit:
    def __init__(self, num_arms, alpha=1.0, beta=1.0):
        self.num_arms = num_arms
        self.alpha = [alpha] * num_arms
        self.beta = [beta] * num_arms

    def select_arm(self):
        samples = [random.betavariate(a, b) for a, b in zip(self.alpha, self.beta…

12 0 Open

A/B testing & experimentation medium

UCB1 Bandit Algorithm in Python

This code implements the UCB1 multi-armed bandit algorithm, balancing exploration and exploitation to identify the best arm while maximizing cumulative reward.

ucb1 bandit ab-testing

Python

import math
import random


def ucb1(means, n_iterations=1000, exploration_weight=2.0):
    """Run UCB1 bandit algorithm on arms with given true means."""
    n_arms = len(means)
    counts = [0] * n_arms
    rewards = [0.0] * n_arms
    
    for t in range(1, n_iterations + 1):
        # UCB1 selection
        if t <…

14 0 Open

Browse by section

Each section groups closely related Python snippets.