autotrain(continuous-openui-local): L01 accept generate_batch_size screening knob - #1438
autotrain(continuous-openui-local): L01 accept generate_batch_size screening knob#1438Tyler-R-Kendrick wants to merge 1409 commits into
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
Important Review skippedToo many files! This PR contains 110 files, which is 10 over the limit of 100. To get a review, reduce the PR to 100 files or fewer by splitting it into smaller PRs or changing its base branch. Upgrade to a paid plan to raise the limit. This review couldn't start because sufficient usage credits or metered capacity aren't available. Add credits or update usage-based reviews in the billing tab, then retry. ⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (110)
You can disable this status message by setting the ✨ Finishing Touches 💡 1🛠️ Fix failing CI checks 💡
📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
When continuous thrash blocks on unacked repair_harness for AgentV/compiler_ms timeouts, the driver now stamps a post-integration receipt commit, acks the action, and rewrites freeze-replay to next_experiment with decode residual priority so soft_clear cannot thrash forever on the same prereq.
Evidence commit for continuous campaign continuous-loop-20260805-continuous-openui-local-8c0b60dd-c46 after integration; enables in-driver ack of decode-timeout repair_harness without human re-prompt.
Closed — live continuous worktree branch, not a landable stack layer
Stack closeout: backlog cleanup 2026-08-06. |
Summary
generate_batch_size=1(fix(autotrain): suite-gate + screening batch-size + promote-block feedback #1433 / autotrain(continuous-openui-local): eval decode-chunk batch-size fix + 3 documented screening cycles (session ixpohr) #1408) butExperimentKnobsforbade the field, so everyHypothesisMatrixvalidation failed before train/eval (c1).evaluate_model --generate-batch-size, bumpharness.autoresearch.experiment_campaignv180 → v181, and regress the matrix + CLI path.insufficient_n).Campaign / cycle
continuous-openui-localgenerate_batch_sizeextra_forbiddendocs/design/continuous-openui-local-c2-canvas-null.mddocs/design/autotrain-cycle-c1-generate-batch-size-schema-failure.mdWhy positive
Harness path error removed; identical continuous measurement path then completes. Fixture ship-gate fails remain soft (not a stack reason for the experiment).
Test plan
pytestmatrix validate + compile_commands generate_batch_size routingverify_version_stamps --check