benchmark-runner

Automate image and video augmentation benchmarks by using configuration-driven YAML files to orchestrate single-library or multi-library comparisons, validating results and generating standardized cross-library reports automatically.

Yes, you can run augmentation library performance comparisons on both local and cloud environments by applying standardized YAML configurations and the benchmark CLI workflow to ensure repeatable experiments.

Validate benchmark results and update documentation automatically by enforcing standardized result structures through Python tooling like pyperf, which triggers report generation and documentation updates after each run.

Run a micro CPU benchmark for RGB image augmentations by executing a local benchmark using a predefined YAML configuration, leveraging the benchmark CLI to orchestrate the test and output standardized performance metrics.

Yes, YAML configs are required to drive multi-library augmentation comparisons, providing the standardized configuration needed to orchestrate runs, validate outputs, and integrate seamlessly with the benchmark CLI workflow.