fix: nautilus_plotter.py runs a real capped search instead of test mode - #150
Merged
Merged
Conversation
Under the release profile (PYAUTO_TEST_MODE=1) Nautilus stops after its first batch with one non-zero weight, so the script's range=0.999 collapsed corner's range to a sliver and corner raised "'range' is not valid or the sample is empty" (Heart release-integrate run 33177898708). The plot example needs a real posterior, so the script declares ENV: real_search (releases PYAUTO_TEST_MODE for this script only) and caps the search with n_like_max=3000, as other workspace examples do. Calibration (release profile, output wiped): 300/500/1000 calls still crash (exploration phase, N_eff 1); 2500 renders but at N_eff 13-20; 3000 gives N_eff 357-527 in 22-26 s (3/3); natural finish is 3500 calls / 35 s. Notebook regenerated with autohands generate.py (the __Env__ section is stripped from it). Upstream: PyAutoLabs/PyAutoFit#1542 (ESS corner guard). Closes #149 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KeLmyZD5aG6BrTTdJzSmDv
This was referenced Aug 28, 2026
Merged
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
scripts/plot/nautilus_plotter.pyfailed under PyAutoHeart's release profile: withPYAUTO_TEST_MODE=1Nautilus stops after its first batch with a single non-zero weight, so the script'srange=0.999collapsed corner's range to a sliver and corner raisedValueError: ... 'range' is not valid or the sample is empty(run 33177898708, jobintegrate / run_scripts (3.12, autofit, plot)).A plot example needs a real posterior, and the human declined raising the global test-mode budget in PyAutoFit (it would slow every release-wave search). So this script now declares
ENV: real_searchin its__Env__section — the runner releasesPYAUTO_TEST_MODEfor this one script — and caps the search withn_like_max=3000, the same lever other workspace examples use.Calibration under the release profile with
output/wiped: 300 / 500 / 1000 calls still crash (Nautilus still in its exploration phase,N_eff = 1); 2000 renders no corner (N_eff = 3); 2500 renders but atN_eff13–20; 3000 givesN_eff357–527 in 22–26 s (3/3 stable); natural finish is 3500 calls / 35 s.emcee_plotter.pystill runs under test mode (control, exit 0).The notebook is regenerated (
generate.pystrips the__Env__section). Corrective PR for the Heart RED reasonrelease validation FAILED (stage integrate).Scripts Changed
scripts/plot/nautilus_plotter.py—n_like_max=3000onaf.Nautilus;__Env__ (Developer Only)section withENV: real_searchnotebooks/plot/nautilus_plotter.ipynb— regeneratedUpstream PR
PyAutoLabs/PyAutoFit#1542
Test Plan
nautilus_plotter.pyexit 0 underprofile_release.yaml, corner rendered, 3/3 runsemcee_plotter.pyunchanged under test mode (exit 0)PYAUTO_TEST_MODEabsent for this script, still1for its siblingsautofit / plotjob greenCloses #149
Generated by the PyAutoLabs agent workflow.
🤖 Generated with Claude Code
https://claude.ai/code/session_01KeLmyZD5aG6BrTTdJzSmDv