Skip to content

google-ml-infra/bap

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

172 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

BAP (Benchmarking Automation Platform)

What is BAP?

BAP is a GitHub-native automation platform designed to standardize performance benchmarking and regression testing for open-source ML frameworks.

At its core, BAP provides a unified system for defining, executing, and tracking benchmarks. It replaces bespoke tooling with a standardized loop:

  • Define: Declare benchmarks in a simple .pbtxt registry.
  • Run: Execute workloads (Python, Bazel, etc.) using a reusable GitHub Actions workflow.
  • Collect: Automatically parse and standardize metrics from TensorBoard logs.

Key Features

BAP provides various capabilities for performance management:

  • Static Threshold Analysis: Automatically compares metrics against a baseline and fails CI jobs if a regression is detected.
  • A/B Testing: A dedicated mode for head-to-head performance comparisons in presubmit to isolate noise and detect regressions accurately.
  • Downstream Consumer Integration: Publishes results via Pub/Sub to for consumption.

Getting Started

  • Onboarding Guide: Step-by-step guide for setting up a benchmark registry and configuring workflows using BAP.

About

GitHub-native automation platform designed to standardize performance benchmarking and regression testing for open-source ML frameworks.

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors