experiment-dse

Automate hyperparameter tuning by supplying an experiment-plan or parameter ranges alongside a config budget. The system generates a configs.yaml list, coordinates batched execution, and compiles a comprehensive DSE report with the best configuration and sensitivity analysis.

Design-space exploration in ML research evaluates multiple hyperparameter configurations to discover optimal setups efficiently. It generates a YAML-dictated config list from parameter ranges, runs batched experiments within budget constraints, and produces a final sensitivity analysis report.

Run batched ML experiments within budget constraints by providing an experiment-plan and a config.yaml budget. The workflow generates parameter configurations, triggers downstream execution tasks via experiment-run, and compiles the final results into a structured report.

Yes, generate a config list from parameter ranges for ML experiments. Supplying parameter ranges and a config budget allows the system to automatically dictate and output a structured configs.yaml file for downstream batched execution and sensitivity analysis.

Compile a sensitivity analysis report by orchestrating end-to-end design-space exploration. After generating configurations and running batched experiments via experiment-run, the system automatically produces a comprehensive DSE report highlighting the best config and parameter sensitivities.