Skip to main content

Benchmark framework for scientists

Project description

Metrika is a library to run and measure benchmarks with scientific rigor. By automating the process of benchmark running, scientists can avoid common mistakes. Metrika allows to obtain statistically solid measures without accidentally introducing methodological errors. In order to do that, the library executes the benchmarks many times and presents results taking into account things like expected value, standard deviation, etc.

This version requires Python 3 or later

Project details


Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page