Skip to main content

Efficient and generalized blackbox optimization (BBO) system

Project description

license Build Status Issues Bugs Pull Requests Version Join the chat at Documentation Status

OpenBox: Generalized and Efficient Blackbox Optimization System.

OpenBox is an efficient and generalized blackbox optimization (BBO) system, which owns the following characteristics:

  1. Basic BBO algorithms.
  2. BBO with constraints.
  3. BBO with multiple objectives.
  4. BBO with transfer learning.
  5. BBO with distributed parallelization.
  6. BBO with multi-fidelity acceleration.
  7. BBO with early stops.

Deployment Artifacts

Standalone Python package.

Users can install the released package and use it using Python.

Distributed BBO service.

We adopt the "BBO as a service" paradigm and implement OpenBox as a managed general service for black-box optimization. Users can access this service via REST API conveniently, and do not need to worry about other issues such as environment setup, software maintenance, programming, and optimization of the execution. Moreover, we also provide a Web UI, through which users can easily track and manage the tasks.


  • Ease of use. Minimal user configuration and setup, and necessary visualization for optimization process.
  • Performance standards. Host state-of-the-art optimization algorithms; select proper algorithms automatically.
  • Cost-oriented management. Give cost-model based suggestions to users, e.g., minimal machines or time-budget.
  • Scalability. Scale to dimensions on the number of input variables, objectives, tasks, trials, and parallel evaluations.
  • High efficiency. Effective use of parallel resource, speeding up optimization with transfer-learning, and multi-fidelity acceleration for computationally-expensive evaluations.
  • Data privacy protection, robustness and extensibility.


Benchmark Results

Single-objective problems Ackley-4 | Hartmann :-------------------------:|:-------------------------: |

Single-objective problems with constraints Mishra | Keane-10 :-------------------------:|:-------------------------: |

Multi-objective problems

DTLZ1-6-5 ZDT2-3

Multi-objective problems with constraints



Installation via pip

For Windows and Linux users, you can install by

pip install lite-bo

For macOS users, you need to install pyrfr correctly first, and then pip install lite-bo.

The tips for installing pyrfr on macOS is here.

Manual installation from the github source

git clone && cd lite-bo
cat requirements.txt | xargs -n 1 -L 1 pip install
python install

macOS users still need to follow the tips to install pyrfr correctly first.

Quick Start

import numpy as np
from litebo.utils.start_smbo import create_smbo

def branin(x):
    xs = x.get_dictionary()
    x1 = xs['x1']
    x2 = xs['x2']
    a = 1.
    b = 5.1 / (4. * np.pi ** 2)
    c = 5. / np.pi
    r = 6.
    s = 10.
    t = 1. / (8. * np.pi)
    ret = a * (x2 - b * x1 ** 2 + c * x1 - r) ** 2 + s * (1 - t) * np.cos(x1) + s
    return {'objs': (ret,)}

config_dict = {
    "optimizer": "SMBO",
    "parameters": {
        "x1": {
            "type": "float",
            "bound": [-5, 10],
            "default": 0
        "x2": {
            "type": "float",
            "bound": [0, 15]
    "advisor_type": 'default',
    "max_runs": 90,
    "time_limit_per_trial": 5,
    "logging_dir": 'logs',
    "task_id": 'hp1'

bo = create_smbo(branin, **config_dict)
inc_value = bo.get_incumbent()
print('BO', '=' * 30)

Releases and Contributing

OpenBox has a frequent release cycle. Please let us know if you encounter a bug by filling an issue.

We appreciate all contributions. If you are planning to contribute any bug-fixes, please do so without further discussions.

If you plan to contribute new features, new modules, etc. please first open an issue or reuse an existing issue, and discuss the feature with us.

To learn more about making a contribution to OpenBox, please refer to our How-to contribution page.

We appreciate all contributions and thank all the contributors!


Related Projects

Targeting at openness and advancing AutoML ecosystems, we had also released few other open source projects.

  • VocalnoML : an open source system that provides end-to-end ML model training and inference capabilities.


The entire codebase is under MIT license

Project details

Download files

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

Files for lite-bo, version 0.6.6
Filename, size File type Python version Upload date Hashes
Filename, size lite_bo-0.6.6-py3-none-any.whl (211.9 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size lite-bo-0.6.6.tar.gz (145.0 kB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page