Repository navigation
Add fresh repeatability, batching and measurement-quality diagnostics to DSPy study - #29
Merged
Merged
Conversation
…solated-cost comparisons with independent raw audits
…eness before fresh multi-call and repeatability diagnostics
CompleteDotTech
added
reproducibility
CompleteDotTech
changed the base branch from
study/dspy-calibration-benchmark
to
main
September 28, 2026 19:32
CompleteDotTech
marked this pull request as ready for review
September 28, 2026 19:32
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Scope
Adds the completion and diagnostic layer to #28: independent replay from eight primary captures, transfer, three service repeatability passes on the original panels, matched batching and isolated requests, data-quality flags, and the original graph replay. It reuses frozen proposals and preserved calls; no new inference was made during this remediation.
The historical run 35345560822 produced raw captures but its analysis failed on a scalar-temperature near-tie, so its downstream audit was skipped. The offline repair keeps that failed status in provenance, bounds the numerical tie explicitly, and rejects material prediction changes. Eight primary captures, eight transfer shards, two repeatability captures, quality sensitivity, and 528 graph replays now validate from the original artifact bytes. The correction does not turn the old hosted run green.
The complete audited publication and durable raw journals follow in #30. New workflows remain manual-only and paid inference jobs remain disabled.