`bayesianbandits`

Bayesian Multi-Armed Bandits for Python

Problem: Despite having a conceptually simple interface, putting together a multi-armed bandit in Python is a daunting task.

Solution: bayesianbandits is a Python package that provides a simple interface for creating and running Bayesian multi-armed bandits. It is built on top of scikit-learn and scipy, taking advantage of conjugate priors to provide fast and accurate inference.

While the API is still evolving, this library is already being used in production for marketing optimization, dynamic pricing, and other applications. Are you using bayesianbandits in your project? Let us know!

Features

Simple API: bayesianbandits provides a simple interface - most users will only need to call pull and update to get started.
Hybrid bandits with cross-arm learning: Share knowledge across similar arms for faster learning and better sample efficiency.
Fast: bayesianbandits is built on top of already fast scientific Python libraries, but, if installed, will also use SuiteSparse to further speed up matrix operations on sparse matrices. Handling tens or even hundreds of thousands of features in a sparse model is no problem.
sklearn pipeline integration: Use sklearn pipelines and transformers to preprocess data before feeding it into your bandit.
Adversarial bandits (EXP3A): Robust performance in non-stationary and adversarial environments.
Flexible: Pick from a variety of policy algorithms, including Thompson sampling, upper confidence bound, and epsilon-greedy. Pick from a variety of prior distributions, including beta, gamma, normal, and normal-inverse-gamma.
Extensible: bayesianbandits provides simple interfaces for creating custom policies and priors.
Well-tested: bayesianbandits is well-tested, with nearly 100% test coverage.

Compatibility

bayesianbandits is tested with Python 3.10, 3.11, 3.12, 3.13, and 3.14 with scikit-learn 1.5.2, 1.6.1, 1.7.2, 1.8.0.

Requires NumPy >= 2.0 and SciPy >= 1.14. For CHOLMOD sparse support, requires scikit-sparse >= 0.5.0, which in turn requires SuiteSparse >= 7.4.0.

Getting Started

Install this package from PyPI.

pip install -U bayesianbandits

Define a LinearUCB contextual bandit with a normal prior.

import numpy as np
from bayesianbandits import (
    Arm,
    NormalInverseGammaRegressor,
    ContextualAgent,
    UpperConfidenceBound,
)

arms = [
    Arm(1, learner=NormalInverseGammaRegressor()),
    Arm(2, learner=NormalInverseGammaRegressor()),
    Arm(3, learner=NormalInverseGammaRegressor()),
    Arm(4, learner=NormalInverseGammaRegressor()),
]

policy = UpperConfidenceBound(alpha=0.84)

Instantiate the agent and pull an arm with context.

agent = ContextualAgent(arms, policy)

context = np.array([[1, 0, 0, 0]])

# Can be constructed with sklearn, formulaic, patsy, etc...
# context = formulaic.Formula("1 + article_number").get_model_matrix(data)
# context = sklearn.preprocessing.OneHotEncoder().fit_transform(data)

agent.pull(context)

Update the bandit with the reward.

agent.update(context, np.array([15.0]))

For shared learning across arms via the design matrix:

from bayesianbandits import LipschitzContextualAgent, ArmColumnFeaturizer, NormalRegressor

# Single shared learner across all arms
agent = LipschitzContextualAgent(
    arms=[Arm(i) for i in range(100)],  # 100 arms sharing knowledge
    learner=NormalRegressor(),
    arm_featurizer=ArmColumnFeaturizer(column_name='article_id'),
    policy=ThompsonSampling()
)

That's it! Check out the documentation for more examples.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

bayesianbandits-1.4.0.tar.gz (336.4 kB view details)

Uploaded Aug 1, 2026 Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

bayesianbandits-1.4.0-cp314-cp314-win_amd64.whl (320.8 kB view details)

Uploaded Aug 1, 2026 CPython 3.14Windows x86-64

bayesianbandits-1.4.0-cp314-cp314-musllinux_1_2_x86_64.whl (739.1 kB view details)

Uploaded Aug 1, 2026 CPython 3.14musllinux: musl 1.2+ x86-64

bayesianbandits-1.4.0-cp314-cp314-musllinux_1_2_aarch64.whl (721.9 kB view details)

Uploaded Aug 1, 2026 CPython 3.14musllinux: musl 1.2+ ARM64

bayesianbandits-1.4.0-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (737.4 kB view details)

Uploaded Aug 1, 2026 CPython 3.14manylinux: glibc 2.17+ x86-64manylinux: glibc 2.28+ x86-64

bayesianbandits-1.4.0-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl (729.0 kB view details)

Uploaded Aug 1, 2026 CPython 3.14manylinux: glibc 2.17+ ARM64manylinux: glibc 2.28+ ARM64

bayesianbandits-1.4.0-cp314-cp314-macosx_11_0_arm64.whl (330.1 kB view details)

Uploaded Aug 1, 2026 CPython 3.14macOS 11.0+ ARM64

bayesianbandits-1.4.0-cp313-cp313-win_amd64.whl (320.0 kB view details)

Uploaded Aug 1, 2026 CPython 3.13Windows x86-64

bayesianbandits-1.4.0-cp313-cp313-musllinux_1_2_x86_64.whl (743.2 kB view details)

Uploaded Aug 1, 2026 CPython 3.13musllinux: musl 1.2+ x86-64

bayesianbandits-1.4.0-cp313-cp313-musllinux_1_2_aarch64.whl (723.0 kB view details)

Uploaded Aug 1, 2026 CPython 3.13musllinux: musl 1.2+ ARM64

bayesianbandits-1.4.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (742.0 kB view details)

Uploaded Aug 1, 2026 CPython 3.13manylinux: glibc 2.17+ x86-64manylinux: glibc 2.28+ x86-64

bayesianbandits-1.4.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl (730.3 kB view details)

Uploaded Aug 1, 2026 CPython 3.13manylinux: glibc 2.17+ ARM64manylinux: glibc 2.28+ ARM64

bayesianbandits-1.4.0-cp313-cp313-macosx_11_0_arm64.whl (329.4 kB view details)

Uploaded Aug 1, 2026 CPython 3.13macOS 11.0+ ARM64

bayesianbandits-1.4.0-cp312-cp312-win_amd64.whl (320.1 kB view details)

Uploaded Aug 1, 2026 CPython 3.12Windows x86-64

bayesianbandits-1.4.0-cp312-cp312-musllinux_1_2_x86_64.whl (744.7 kB view details)

Uploaded Aug 1, 2026 CPython 3.12musllinux: musl 1.2+ x86-64

bayesianbandits-1.4.0-cp312-cp312-musllinux_1_2_aarch64.whl (726.3 kB view details)

Uploaded Aug 1, 2026 CPython 3.12musllinux: musl 1.2+ ARM64

bayesianbandits-1.4.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (744.9 kB view details)

Uploaded Aug 1, 2026 CPython 3.12manylinux: glibc 2.17+ x86-64manylinux: glibc 2.28+ x86-64

bayesianbandits-1.4.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl (736.1 kB view details)

Uploaded Aug 1, 2026 CPython 3.12manylinux: glibc 2.17+ ARM64manylinux: glibc 2.28+ ARM64

bayesianbandits-1.4.0-cp312-cp312-macosx_11_0_arm64.whl (330.1 kB view details)

Uploaded Aug 1, 2026 CPython 3.12macOS 11.0+ ARM64

bayesianbandits-1.4.0-cp311-cp311-win_amd64.whl (320.0 kB view details)

Uploaded Aug 1, 2026 CPython 3.11Windows x86-64

bayesianbandits-1.4.0-cp311-cp311-musllinux_1_2_x86_64.whl (748.0 kB view details)

Uploaded Aug 1, 2026 CPython 3.11musllinux: musl 1.2+ x86-64

bayesianbandits-1.4.0-cp311-cp311-musllinux_1_2_aarch64.whl (734.8 kB view details)

Uploaded Aug 1, 2026 CPython 3.11musllinux: musl 1.2+ ARM64

bayesianbandits-1.4.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (744.0 kB view details)

Uploaded Aug 1, 2026 CPython 3.11manylinux: glibc 2.17+ x86-64manylinux: glibc 2.28+ x86-64

bayesianbandits-1.4.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl (739.1 kB view details)

Uploaded Aug 1, 2026 CPython 3.11manylinux: glibc 2.17+ ARM64manylinux: glibc 2.28+ ARM64

bayesianbandits-1.4.0-cp311-cp311-macosx_11_0_arm64.whl (329.3 kB view details)

Uploaded Aug 1, 2026 CPython 3.11macOS 11.0+ ARM64

bayesianbandits-1.4.0-cp310-cp310-win_amd64.whl (320.0 kB view details)

Uploaded Aug 1, 2026 CPython 3.10Windows x86-64

bayesianbandits-1.4.0-cp310-cp310-musllinux_1_2_x86_64.whl (722.8 kB view details)

Uploaded Aug 1, 2026 CPython 3.10musllinux: musl 1.2+ x86-64

bayesianbandits-1.4.0-cp310-cp310-musllinux_1_2_aarch64.whl (708.0 kB view details)

Uploaded Aug 1, 2026 CPython 3.10musllinux: musl 1.2+ ARM64

bayesianbandits-1.4.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (719.2 kB view details)

Uploaded Aug 1, 2026 CPython 3.10manylinux: glibc 2.17+ x86-64manylinux: glibc 2.28+ x86-64

bayesianbandits-1.4.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl (714.7 kB view details)

Uploaded Aug 1, 2026 CPython 3.10manylinux: glibc 2.17+ ARM64manylinux: glibc 2.28+ ARM64

bayesianbandits-1.4.0-cp310-cp310-macosx_11_0_arm64.whl (329.8 kB view details)

Uploaded Aug 1, 2026 CPython 3.10macOS 11.0+ ARM64

File details

Details for the file bayesianbandits-1.4.0.tar.gz.

File metadata

Download URL: bayesianbandits-1.4.0.tar.gz
Upload date: Aug 1, 2026
Size: 336.4 kB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: uv/0.12.1 {"installer":{"name":"uv","version":"0.12.1","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

File hashes

Hashes for bayesianbandits-1.4.0.tar.gz
Algorithm	Hash digest
SHA256	`99883725684a4f20cca16263082e017c9d69742f17474b20e9dce4dcabb09c41`
MD5	`7b31ee8d28646b4c33cc0a5b753f4e2e`
BLAKE2b-256	`4f1117d99075bdd4dac79007350a538e13ebdcfe80779422716077e1d9681148`

Algorithm	Hash digest
SHA256	`3b9be53ef51d88a94e37eed1fe2361ef7ce43d59f63f50ea37b23f8e3d2e3541`
MD5	`f5dbf225d852d85bf43329980a13d456`
BLAKE2b-256	`f2a2eec993e9092bd2b77aa293976ab6ea938a1d7d60a16e0f61166b68f5c914`

Algorithm	Hash digest
SHA256	`9322b97c7519325dd0d8a5b99c287e74cfeb5caef4a371ba3cf04c8356f60eea`
MD5	`d5da48e93055bed72b660627ae963744`
BLAKE2b-256	`d8ea897f2962088fbec95610b76511e147aadf57782cd32cf73e9cb254100104`

Algorithm	Hash digest
SHA256	`74f56ee8f0ed16e536bd833690f5537a64abd71c0db1e838e7f10c84c85534a7`
MD5	`b9327d6a1022401c214c30b788c4132c`
BLAKE2b-256	`b94c080451a50f55073cdabd296b0d64ec9b7ef0ff5a31ddfdd5e5a996884aa4`

Algorithm	Hash digest
SHA256	`10cd8ac5c85b88383bccf5343d9f7a60b46fb3dd8fe9c21ed7fabdba8fc358d2`
MD5	`3e1d17aba630aabae76c4eed3044ae15`
BLAKE2b-256	`a005c52e175fda4edf2194329ad59dd54aa792887d2b29eb3503bcee981bc1f8`

Algorithm	Hash digest
SHA256	`7a4ea8c5c422975c36140fb5828ee29778698b11e4ca88fff2f69920902c7eab`
MD5	`da6de3f5ce27da492f05231d365c4e28`
BLAKE2b-256	`6412195398c9823c8748466a4ab590404cfe23fe24091f246bff8056a1078c24`

Algorithm	Hash digest
SHA256	`bce74e803a4f797ae415ba0218a0ff20a5cf7336c7cf18c6b35db0428f813a4a`
MD5	`92616633c6745b7fa845c11bcf17e06b`
BLAKE2b-256	`0733531c5ad00787f90b1d94f95b8a900f360b274306d714c3267ec8b4704d27`

Algorithm	Hash digest
SHA256	`9ff5a9bbc3cb106bc7ff2b0d0261b37cc85da6e153641e5a4dcb5b042b8f71e5`
MD5	`037d2c5674c62801d40854ca042fd487`
BLAKE2b-256	`96e8543465f69d22dcd73366d0969c249da4f9a1751067ae463ccbed0db388c8`

Algorithm	Hash digest
SHA256	`f3e6ae4e67d96dd6f25223669c15dc99a4a12be80106c4337edef56ebeca159c`
MD5	`4cfbfde82097397928e27084e5fd7d6e`
BLAKE2b-256	`d78f920daed894f90fcc05dc3c474f20a31178b475b253bdf4ef3404f80225db`

Algorithm	Hash digest
SHA256	`21f8e1161b3b1e52435b7bbdf620852acd547056c98aaf415f5b1b9965f1caaf`
MD5	`eba0bc725d080d599a767dfe165706be`
BLAKE2b-256	`b25c802ee6457497234228cdc71b74f22fff9e0df7fb1b0115d8ed81179b0e08`

Algorithm	Hash digest
SHA256	`9a18fa656a0b70ab081cbe269e6bef482fb0db04e0717e08ea3e34b9d26f81d7`
MD5	`44579923efcf8e58b0a5114ca77abe5b`
BLAKE2b-256	`e893d8b5af8463193f0e41022ba2b33388d3d13a9cbb8b608264244b325360c0`

Algorithm	Hash digest
SHA256	`2bef0269530003e96198c27da8e24a812223def85cc032db484dd3fc7364fbbf`
MD5	`0da6b290e092212682a676c0bfb68759`
BLAKE2b-256	`1521972df170a579dac63e2bfffb5e8fb1af26de1f4a35cc4440759441e90b3f`

Algorithm	Hash digest
SHA256	`0db476ca4456d43a4d122eca6dea30c94fe5d5f7c5623c4fc804d0fc2b2a6c50`
MD5	`43dd034cdc08829d25b5e23098454d31`
BLAKE2b-256	`140b8b81afd9352fdc0a22d652a5967815b84e5ba4852fdfeb668bcb8d21b9e9`

Algorithm	Hash digest
SHA256	`07d0e325b871fce43f17ac99220d6a5b7608bf4dff2276d3b1e419da791b261d`
MD5	`87d18c9cce45dd04c5cab592619da703`
BLAKE2b-256	`d171868b7260927e1ab31360b42d93f02ddd09218b0c61863c4fc365482b4e1f`

Algorithm	Hash digest
SHA256	`6ada6ce95dfded26d55ba6f8f38a8d6e2586b67944516f9cbdebaf610baff521`
MD5	`7681b8d91130dc0ced8236e6b3e31477`
BLAKE2b-256	`cc559e2afef7414c38cd91b2a5e9e85ed0e0bfa200fbd4c086c523291dfd74ae`

Algorithm	Hash digest
SHA256	`d449fea20176da7e9c09487f781493773f3513b59823960c1726efe196b89d04`
MD5	`0a9b2951ea72773ae5cdedb8c0f43dfc`
BLAKE2b-256	`987d8aa0680be227273e4710546a9e1ffaf636a5927a381fed2908fcc2229ba3`

Algorithm	Hash digest
SHA256	`fd210cc10f497b58c0e3167de8f38df13cb2c59a1581e267654288a859174a94`
MD5	`122b78822ff7d363c455fea2ba004288`
BLAKE2b-256	`42f4c308da1b72bea7002945006b47560089841a84d3402353fbf904218d7ef6`

Algorithm	Hash digest
SHA256	`f6ced694745438c8373484fa1ae791a4c93ed5fd5a2730be8739614da66a1e3c`
MD5	`9917e583554a0b6754a75ade48dc6160`
BLAKE2b-256	`d088b15fb280af12287066012fa4e27f24141562f18a7ada6f36606f1a103dea`

Algorithm	Hash digest
SHA256	`858a901415e1d61e3d6adb72b185d9d8054ee907c9fa933bacbacd99bf93e4c6`
MD5	`45a5bb0fe26b968cb2d8f4f58db51795`
BLAKE2b-256	`5f92e5e841e6302f2b8fe6463887b07e8528b92b3e280bdd42b38b13518c93c7`

Algorithm	Hash digest
SHA256	`6925f01e8fb18130698989144ba62de904eb41c3dc795718482522cf9a02572c`
MD5	`203d2babe538addc579d7bba4b49b7f6`
BLAKE2b-256	`8b6bd14a4142870ff2bec67dc3453a065798d3c8cceb0eaac2045ad0c928a3b2`

Algorithm	Hash digest
SHA256	`1afd1bf8219367fad63489db94be8c212dcff02c2ce50222ea8f06e49940cbd1`
MD5	`ada7d4c9e25f6effd8deac476821d1ea`
BLAKE2b-256	`217aaa8c5a98e46ab4d9c45c21fba6b2c4fb233f44a6bbb486f65a6657cf6b20`

bayesianbandits 1.4.0

bayesianbandits

Bayesian Multi-Armed Bandits for Python

Features

Compatibility

Getting Started

Key dates

1 maintainer

Credits

Requires

Provides Extra

Download files

Source Distribution

Built Distributions

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

`bayesianbandits`

Algorithm	Hash digest
SHA256	`7a7d63789f9595aabdc1c7e315366c3123c782e92caa0d39164aef9f20d41234`
MD5	`b77a8958cff82f1dc8f360ede9cd772f`
BLAKE2b-256	`4b080be697ff2aa97b96878adf4d8a0674f2c28147cc9b0ff6cc5544edf2d50f`

Algorithm	Hash digest
SHA256	`3de05dc5b2ab66b117476be14e5b6afd49036e1b98cc6b482d358f38064c85b4`
MD5	`8f177144eca628f015bb801d285e0d43`
BLAKE2b-256	`c42be4a928e10e2fa5cb83069ed91851deb4371707aa5e65d3a0e17d4214b15a`

Algorithm	Hash digest
SHA256	`c3a547b16da9af4e8f88d81cbab451f2cd0a37d26e22821ceddaf03861c6c355`
MD5	`20d939e1ec4cb7d1fde1dc014ce52269`
BLAKE2b-256	`a8360d1fe64d2b234209cfd046cbf8bd06055c8764845e477885b312ad215dd4`

Algorithm	Hash digest
SHA256	`e2872b56dab3ea6c9909c0f96b30bf5f913212fb80afa7d7417aecc649c9b868`
MD5	`b45592c71fd4a3befdadb0c6734246a6`
BLAKE2b-256	`ec07229142190b33ab88bb14d2b2c48d58cbe234c2bd3f67f47196ad0158f642`

Algorithm	Hash digest
SHA256	`ad646948ef0e01c91d4ca721ae6aad3c7830259de3f1fdf8d0e598043b84bb73`
MD5	`425ba1ef329d197af25e6f8d5dd71575`
BLAKE2b-256	`7e3975d6662bb343e4315ebaf2117473af14ff89ef5ede42bffef5e6075a4f36`

Algorithm	Hash digest
SHA256	`699db60e3ca4944103ec6233f53eb5fc4f479ed2330c01da2a655cf86890e7fa`
MD5	`c91bb9a047e95acd41ca33cf677b2d91`
BLAKE2b-256	`b12a1676d3705eca06175c202981d8579b9bcff27ec4944eb32435cdbceb15b6`

Algorithm	Hash digest
SHA256	`ad083d0a390dd51cb75b96933b819b1da5333c55ff0ac2a2d84c0b73a60090c3`
MD5	`7333384c2be4e48b17b41d714e07512b`
BLAKE2b-256	`058f0c2df33f2ad4e8e6637cb019fb8aa94b9e134b4650b2df1c69e3806e4496`

Algorithm	Hash digest
SHA256	`cd72c7d8c927d6b97462ea26f142d62dbba464986111cbee5cc37b725767f075`
MD5	`437525be586127bc70d5866b779a0f96`
BLAKE2b-256	`4f66b60245ef80118536d1e28d95596567a2c71712086e85f232c6024df3741c`

Algorithm	Hash digest
SHA256	`4c16e704a261ef99380f3606398f35f167981f4e72188cad051dac4aff3b12ff`
MD5	`799ba0abf0f7e767363a15db5c88fb97`
BLAKE2b-256	`d83d86ed19881370562e2621dfefa74b050262fc95c9163461ca11a945863322`

Algorithm	Hash digest
SHA256	`cf7d5ad417f14365be7d67d8522c7d940eefbb79047a1bd45cc2e6af5b068c0c`
MD5	`25b10ab785973ea30bd8a38e22ad8a24`
BLAKE2b-256	`80c58762ce1088f1d88b4822117b10324a9288e80deab85127a16534d1f1cd90`