explore-run
Rank small training trials before bigger runs.
What it does
Plans bounded exploratory training trials and produces ranked candidate summaries with fair-comparison caveats.
When to use it
Use when the researcher explicitly authorizes exploratory trials and wants candidate rankings rather than certified success.
How to use it
Give it an explicitly authorized exploratory trial task in a research repository; it writes candidate rankings and comparison reports into explore_outputs/.
What you provide
- An existing project
What you need · 4
Requires Python 3 to run the bundled planning and output scripts.
Requires its shared operating principles for candidate run planning.
May receive actual command execution as an alternative to run-train; installation is not stated as required.
May receive actual command execution as an alternative to minimal-run-and-audit; installation is not stated as required.
Access · 0
No additional actions listed.