Repository navigation
O4c: fair CIP posterior export via systematic resampling, final group unique (port from rift_O4d) - #218
Open
oshaughnessy-junior wants to merge 3 commits into
Conversation
oshaughnessy-junior
deployed
to
private-review-dispatch-rift-upstream
October 12, 2026 11:54 — with
GitHub Actions
Active
oshaughnessy-junior
deployed
to
private-review-dispatch-rift-upstream
October 12, 2026 12:18 — with
GitHub Actions
Active
oshaughnessy-junior
had a problem deploying
to
private-review-dispatch-rift-upstream
October 12, 2026 12:18 — with
GitHub Actions
Error
…ift_O4d) Port of oshaughnessy-junior/research-projects-RIT rift_O4d bce0862, 5e802f3, b8865fc (junior PR #56), CIP and cip_pipeline parts. Weighted np.random.choice(replace=False) is successive sampling, so the first-N truncation biased the export toward the proposal. Systematic resampling on the weight CDF has exact expected counts at any N. Adds --posterior-unique-draw and the +annotation_export.dat sidecar. The pseudo_pipe hunk follows in the next commit. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Ports the pseudo_pipe hunk of rift_O4d bce0862 (oshaughnessy-junior #56): the final args_cip_list.txt line gets --posterior-unique-draw (skipped on the AMR path; a final G line is left alone), and lines are written newline-terminated. Not ported, because rift_O4c lacks what they guard: the --internal-ile-deduplicate-grid exclusion and the MultiApprox CIP_terminal strip (rift_O4d ca30121, oshaughnessy-junior #218). Adds a test that runs the writer block. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…info skip Review [P2] on PR #56: the equal-weights exactness test evaluated np.longdouble(10.0)**400 unconditionally; on platforms where longdouble has only float64 range that is inf, unique_draw_bound correctly raises, and the test fails on platforms RIFT supports. The beyond-DBL_MAX assertion now lives in the extended-precision test after its finfo skip; the np.ones assertions stay unconditional. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> (cherry picked from commit 4633f83)
oshaughnessy-junior
force-pushed
the
claude/o4c-cip-fair-export
branch
from
October 12, 2026 12:18
f45ec8e to
0d2aaca
Compare
oshaughnessy-junior
deployed
to
private-review-dispatch-rift-upstream
October 12, 2026 12:19 — with
GitHub Actions
Active
oshaughnessy-junior
marked this pull request as ready for review
October 12, 2026 12:23
oshaughnessy-junior
deployed
to
private-review-dispatch-rift-upstream
October 12, 2026 12:23 — with
GitHub Actions
Active
This branch was successfully deployed
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.
Ports the fair CIP posterior export from
rift_O4d: oshaughnessy-junior#56 (bce0862, 5e802f3, b8865fc, 4633f83).rift_O4cexports with weightednp.random.choice(replace=False)and keeps the first N, which biases the export toward the proposal. On a GW150914 fixed-archive refit, the q median moved from 0.852 to 0.839.--posterior-unique-draw; an+annotation_export.datsidecar.--posterior-unique-draw. The AMR path and a final G group are skipped.--internal-ile-deduplicate-gridguard and the MultiApproxCIP_terminalstrip (junior O4c: fair CIP posterior export via systematic resampling, final group unique (port from rift_O4d) #218). O4c has neither feature.Tests on ldas-grid with the IGWN CVMFS python, no cupy:
test_cip_pipeline.py,test_rf_transverse_spin.py,test_geometric4.py: 116 passed. The new writer test fails on the unpatched code.CIP_3.sub.rift_O4dfc532d0 seed for seed. The flagged run delivers 5188 distinct rows (the bound).🤖 Generated with Claude Code