Skip to content

Commit cf68f15

Browse files
luisleo526claude
andauthored
docs: use neutral wording for historical performance examples (#728)
Reword the historical changelog example using neutral references, preserving its report-sink and cache conditions and recorded timings while removing the script name and example trade count. This change updates documentation wording only. ## Commits - docs: use neutral wording for historical performance examples Merged by the spark L6 on the exact head 061f6c7 after the engine merge gate passed on the docs-only basis: pineforge/verify (lab remote preflight + docs on tree 4a46db5, job rj-20261103t224900-fd397c; docs built with 14 non-fatal Doxygen warnings) and pineforge/parity (no engine behaviour change: 1 file, CHANGELOG.md), plus every Actions check. Co-Authored-By: Claude Opus 5.5 <[email protected]>
1 parent 8e1fcf3 commit cf68f15

1 file changed

Lines changed: 2 additions & 2 deletions

File tree

‎CHANGELOG.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1411,13 +1411,13 @@ deployments using Docker `--tmpfs /tmp` or Kubernetes `emptyDir` with
14111411
effect. The release image's `docker/run_json.py` default (unjailed)
14121412
profile installs a temporary report sink on every handle, so there the
14131413
cache holds 128 rows and a script that loops over its closed trades reads
1414-
them from scratch files on every pass: hungpixi (24,238 closed trades),
1414+
them from scratch files on every pass: one measured example,
14151415
one run each started together on one 16 vCPU machine, took 74.18 s on
14161416
1.6.3 and did not finish within 1800 s on 1.7.0, where the run was
14171417
stopped at that limit. The trigger is a report sink installed together
14181418
with a loop over more than 128 closed trades. Not affected:
14191419
`scripts/run_strategy.py` without a sink, which the hosted case runner,
1420-
the verifier and the sweep run (hungpixi takes 107.2 s there, mean of
1420+
the verifier and the sweep run (the same example takes 107.2 s there, mean of
14211421
three runs; the corpus-parity canary pass sets
14221422
`PINEFORGE_REPORT_SINK=1`), and the HPO native adapter, which installs no
14231423
sink.

0 commit comments

Comments
 (0)