Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Experiment validation #81

Open
WarmCyan opened this issue Sep 18, 2023 · 2 comments · May be fixed by #93
Open

Experiment validation #81

WarmCyan opened this issue Sep 18, 2023 · 2 comments · May be fixed by #93
Labels
cli enhancement New feature or request

Comments

@WarmCyan
Copy link
Collaborator

The experiment [...] --map command is a good way to quickly (and informally) check that an experiment is structured correctly and for determining which artifacts are cached, but it would be nice if there was a more explicit way to run some set of validations for a given experiment CLI line, to help a user determine if the experiment is going to run correctly before potentially wasting hours on an invalid command.
@WarmCyan WarmCyan added enhancement New feature or request cli labels Sep 18, 2023
@WarmCyan WarmCyan mentioned this issue Sep 18, 2023
Closed
@WarmCyan
Copy link
Collaborator Author

CLI could be --check, and it does the mapping but rather than displaying the map, outputs statuses, number of artifacts (and how many were cached), number of stages to run, number of records etc, essentially as a summary view

@WarmCyan WarmCyan mentioned this issue Sep 22, 2023
Open
@WarmCyan WarmCyan linked a pull request Oct 4, 2023 that will close this issue
Draft
@WarmCyan WarmCyan linked a pull request Oct 4, 2023 that will close this issue
Experiment validation #93
Draft
@WarmCyan
Copy link
Collaborator Author

Should check:

  • Any exceptions from trying to import all the necessary files?
  • Do all of the parameter files return non-empty lists of parameter sets?
  • Are all stage inputs accounted for from previous stage outputs? (e.g. valid map)

Stats: (if --debug is specified, for each of the following actually list out the things)

  • How many parameter files, how many parameter sets from each, how many total parameter sets
  • How many records generated, how many stages per, how many stages total
  • How many artifacts, how many were cached, how many stages actually have to run
  • Has this exact experiment been run before? What are the run ids from that? How many were full store runs? How many passed?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
cli enhancement New feature or request
Projects
Roadmap
Status: WIP
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Experiment validation ORNL/curifactory
1 participant
@WarmCyan