google-vizier 0.0.4

Vizier: Distributed service framework for blackbox optimization and research.

Open Source Vizier: Reliable and Flexible Blackbox Optimization.

Documentation | Installation | Citing Vizier

What is Open Source (OSS) Vizier?

OSS Vizier is a Python-based service for blackbox optimization and research, based on Google Vizier, one of the first hyperparameter tuning services designed to work at scale.

OSS Vizier's distributed client-server system. Animation by Tom Small.

OSS Vizier's interface consists of three main APIs:

• User API: Allows a user to setup an OSS Vizier Server, which can host blackbox optimization algorithms to serve multiple clients simultaneously in a fault-tolerant manner to tune their objective functions.
• Developer API: Defines abstractions and utilities for implementing new optimization algorithms for research and to be hosted in the service.
• Benchmarking API: A wide collection of objective functions and methods to benchmark and compare algorithms.

Please see OSS Vizier's ReadTheDocs documentation for detailed information.

Installation

To install the core API, the simplest way is to run:

pip install google-vizier

which will download the code and launch the provided install.sh. This script installs the necessary dependencies from requirements.txt, and builds the relevant protobuf libraries needed for the service. OSS Vizier requires Python 3.10+, while client-only packages require Python 3.7+.

For full installation (in particular to support all algorithms and benchmarks), you will need to install all dependencies via install_extras.sh. This will install the following:

• requirements-jax.txt: Jax libraries shared by both algorithms and benchmarks.
• requirements-tf.txt: Tensorflow libraries shared by both algorithms and benchmarks.
• requirements-algorithms.txt: Additional repositories (e.g. Emukit) for algorithms.
• requirements-benchmarks.txt: Additional repositories (e.g. NASBENCH-201) for benchmarks.

Check if all unit tests work by running run_tests.sh.

Citing Vizier

If you found this code useful, please consider citing the OSS Vizier paper as well as the Google Vizier paper. Thanks!

@inproceedings{oss_vizier,
  author    = {Xingyou Song and
               Sagi Perel and
               Chansoo Lee and
               Greg Kochanski and
               Daniel Golovin},
  title     = {Open Source Vizier: Distributed Infrastructure and API for Reliable and Flexible Blackbox Optimization},
  booktitle = {Automated Machine Learning Conference, Systems Track (AutoML-Conf Systems)},
  year      = {2022},
}
@inproceedings{google_vizier,
  author    = {Daniel Golovin and
               Benjamin Solnik and
               Subhodeep Moitra and
               Greg Kochanski and
               John Karro and
               D. Sculley},
  title     = {Google Vizier: {A} Service for Black-Box Optimization},
  booktitle = {Proceedings of the 23rd {ACM} {SIGKDD} International Conference on
               Knowledge Discovery and Data Mining, Halifax, NS, Canada, August 13
               - 17, 2017},
  pages     = {1487--1495},
  publisher = {{ACM}},
  year      = {2017},
  url       = {https://doi.org/10.1145/3097983.3098043},
  doi       = {10.1145/3097983.3098043},
}