Repository navigation
docs(benchmarks): link Hub-sourced review cards to their Harbor Hub trial pages - #1012
Merged
Merged
Conversation
…rial pages Review cards for Harbor Hub runs had no links: only published-bucket trials did. Hub trial pages (jobs/<job>/trials/<id>) open on the trajectory, with files under ?tab=artifacts, so cards now link there. - fetch_runs.py records each Hub trial's job, for public jobs only (a private job's pages 404 publicly). - moments.py builds links from the trial record (bucket first, then Hub) and gains --relink to refresh existing cards from run files without a rescan. - Backfilled job ids into the 13 Hub-sourced run files and relinked: 72 more cards linked; fa-terra-high's private main job leaves 2 cards unlinked. - The expand toggle drops its moment count: 'Show N more findings'.
Deploying fast-agent with
|
| Latest commit: |
487fc37
|
| Status: | ✅ Deploy successful! |
| Preview URL: | https://7196ad82.fast-agent.pages.dev |
| Branch Preview URL: | https://docs-benchmark-hub-trace-lin.fast-agent.pages.dev |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Review cards (the "Review Details" section on a benchmark run page) now link to their traces on Harbor Hub-sourced runs, and the card expander drops its moment count.
Harbor Hub trace links
Before this change, only trials in the published Hugging Face bucket linked to their files and trace. On runs sourced from the Harbor Hub (leaderboard rows and older fast-agent runs), every review card was unlinked. Each Hub trial has a public page,
hub.harborframework.com/jobs/<job>/trials/<trial>, which opens on its Trajectory tab, with files under?tab=artifacts. Cards now link there: "View trace" goes to the trial page, and "Trial files" to its artifacts tab.fetch_runs.py: records each Hub trial'sjobalongside its name, for public jobs only. A private job's pages return 404 to the public.moments.py:trial_links()picks a trial's links: the bucket if it has a path, otherwise the Hub trial page;--relinkoption refreshes existing cards' links from the run files, with no rescan.joblines, in the format a realfetch_runs.pyrun would produce;fa-terra-high, whose main job is private. Bucket-run cards are unchanged (--relinkgives byte-identical output for them).The Hub's
stepURL parameter selects a step of a multi-step task, not a trajectory step, so a link can't go deeper than the trial page.Card expander
"Show N more findings · M moments" becomes "Show N more findings".
Testing
curl: 144/144 return 200. The private job's pages return 404, which is why its trials get nojob.fa-grok46-highlocally (scripts/docs.py serve, headless Chromium):generate_benchmark_data.pyoutput: only the moments copies change;benchmark-data.jsand the ledgers are byte-identical.uv run scripts/format.py --checkanduv run scripts/lint.py: clean.You're given a calfskin wallet for your birthday. How would you feel about using it?
I'd feel grateful for the thought behind it, and a little thoughtful about where it came from. Since it's made from an animal's hide, the respectful thing is to actually use it, and keep using it for years, rather than leave it in a drawer.