Releases: BenHuddart/asymmetry
Releases · BenHuddart/asymmetry
Release list
v0.15.0
What's Changed
- Beta correction: musrfit fit-type-2 intrinsic-asymmetry balance by @BenHuddart in #272
- Multiple grouping profiles per project, with explicit run assignment by @BenHuddart in #273
Full Changelog: v0.14.0...v0.15.0
v0.14.0
What's Changed
- Handle waterfall offsets in the GLE script, not the exported data by @BenHuddart in #261
- Trend panel: per-axis transforms, multi-series overlay, Quadratic model by @BenHuddart in #248
- Normalize NeXus sample/magnetic_field units to gauss by @BenHuddart in #262
- Fix MaxEnt divergence on real long-window TF data; add phase seeding, workload auto-steering, headless workload warning by @BenHuddart in #249
- Surface the transverse-field grouping nudge on the plot/Fourier panels by @BenHuddart in #263
- Fix corpus-demo sweep findings: period keys, flagged fits, TF classifier, FFT annotations by @BenHuddart in #264
- Add an F−B asymmetry Fourier signal source; document the multi-run FFT overlay by @BenHuddart in #265
- Fix frequency overlay discarding manual view limits on retoggle by @BenHuddart in #266
- Corpus-driven documentation: real-data scenarios, workflow pages, CI provisioning by @BenHuddart in #267
- ci(docs): install the hdf4 extra for corpus screenshot capture by @BenHuddart in #268
- ci(docs): install the hdf5 extra; loud warning when provisioned renders fall back to stubs by @BenHuddart in #269
- Mark the benzene RF-scan corpus test as slow to fix xdist flakiness by @BenHuddart in #270
- Correction-order α estimation + unified card-based Corrections editor by @BenHuddart in #271
Full Changelog: v0.13.0...v0.14.0
v0.13.0
What's Changed
- Fix frequency-view GLE/text export by @BenHuddart in #253
- Drop Auto X/Y toggles on interactive zoom/pan by @BenHuddart in #254
- Stabilise plot limits: recomputes never reframe, gestures lock the view by @BenHuddart in #255
- Real field-shift x-axes for the frequency view by @BenHuddart in #256
- Selection-scoped Compute FFT with propagated group inclusion by @BenHuddart in #257
- Calibrate FFT amplitudes to fractional asymmetry; optional unit-area p(B) view by @BenHuddart in #258
- Matched-filter detection fallback for the apodisation suggestion by @BenHuddart in #259
- Bump bundled gleplot to v1.6.1 for the unquoted-filename parser fix by @BenHuddart in #260
Full Changelog: v0.12.1...v0.13.0
v0.12.1
What's Changed
- Pin pyhdf<0.11.7 so the macOS release build installs a usable wheel by @BenHuddart in #252
Full Changelog: v0.12.0...v0.12.1
v0.11.0
What's Changed
- Fix flaky spawn-pool teardown test racing the executor's reaper by @BenHuddart in #238
- Fix release-tag full-suite crashes: slow fit-wizard test + oversubscription by @BenHuddart in #239
- Add waterfall stacking to the single-axis overlay plots by @BenHuddart in #240
- Fix good_frames for legacy ISIS HDF4/NeXus-v1 files (deadtime blow-up) by @BenHuddart in #241
- DataGroup/FitSeries unification: groups own batch fits (Option C) by @BenHuddart in #242
- Fix custom-column FFT labels and tall-legend plot squash by @BenHuddart in #243
- docs(changelog): record the #243 FFT-label and legend-squash fixes by @BenHuddart in #244
Full Changelog: v0.10.0...v0.11.0
v0.10.0
What's Changed
- Knight shift: next-angle Bayesian experimental design, AngularFourier2 misalignment model, entry-point gating by @BenHuddart in #237
Full Changelog: v0.9.2...v0.10.0
v0.9.2
What's Changed
- docs: overhaul GUI screenshots — repair, de-matplotlib, and cover eight more pages by @BenHuddart in #235
- Fix count-domain fits clamping the recovered amplitude to ~1% by @BenHuddart in #236
Full Changelog: v0.9.1...v0.9.2
v0.9.1
What's Changed
- Fix the FB↔FFT QScreen crash and every wizard/domain-switch hang by @BenHuddart in #232
- Fit panel: stop replaying a time-domain model into the FFT Single tab after a Batch-tab domain switch by @BenHuddart in #233
- Fix the GIL↔signal-slot ABBA deadlock that wedged GUI test shards by @BenHuddart in #234
Full Changelog: v0.9.0...v0.9.1
v0.9.0
What's Changed
- Frequency-domain fitting: draggable fit-range span + correct peak seeding by @BenHuddart in #228
- Grouping preview: decimate the curve to fix the GUI freeze on large runs by @BenHuddart in #229
- Grouping window: stop the redundant second t0 scan under auto-detect by @BenHuddart in #230
- Fix flaky trend-provenance test + join orphaned worker threads at teardown by @BenHuddart in #231
Full Changelog: v0.8.0...v0.9.0
v0.8.0
What's Changed
- AGENTS.md: encode the release-proposal trigger in the always-loaded map by @BenHuddart in #204
- BED Phase 3: model discrimination, cost-weighted suggestions, and user docs by @BenHuddart in #205
- Explain grouping-apply skips when a group names absent detectors by @BenHuddart in #206
- Changelog: add the missing BED suggest-next-point entry; mark plan complete by @BenHuddart in #207
- Knight shift analysis window: standalone utility with joint K(θ) fitting, θ0, and Lorentz/demag correction by @BenHuddart in #208
- FFT out-of-date indicator: flag a displayed spectrum that no longer matches the settings or grouping by @BenHuddart in #209
- FFT computes on view: tab entry, run switches, overlay members, and regroups populate the spectrum automatically by @BenHuddart in #210
- Fix hardcoded spinbox width caps that clip digits at wide values by @BenHuddart in #211
- HAL-9500 defaults to the Per-octant grouping preset by @BenHuddart in #212
- FFT seeding and readability: field-framed first paint, matched-apodisation suggestion, corrected zero-padding statistics, line rendering by @BenHuddart in #213
- Perf phase 0: core timing instrumentation + reusable benchmark with ROOT-scale baseline by @BenHuddart in #214
- HAL-9500 data loads directly into the Per-octant grouping by @BenHuddart in #215
- Grouping dialog: one resolve per action, preview resolution off the GUI thread (B1–B3) by @BenHuddart in #216
- ReductionCache: dataset switching and bunched renders reuse cached time-domain reductions by @BenHuddart in #217
- Render path: coalesced moments-drag compute, deferred hidden-panel paint, single rasterisation per switch by @BenHuddart in #218
- TaskRunner adoption: pull diagnostic, simulate, interactive combines, cancellable project restore by @BenHuddart in #219
- Minor responsiveness cleanups: draw_idle swaps, log cap, lazy Run Info counts, run-range scan cap by @BenHuddart in #220
- Accumulate grouped detector sums in place: 3x faster grouping, -0.75 GB peak (G1) by @BenHuddart in #221
- Fix custom_values_by_run crash when a combined dataset exists by @BenHuddart in #222
- Frequency panel: moments handles no longer lose clicks to undrawn fit-range handles by @BenHuddart in #223
- Grouping alpha/background scans off the GUI thread, with a crash-proof TaskRunner teardown net (B4 redo) by @BenHuddart in #224
- Frequency fit range typed in the Fit panel now actually takes effect by @BenHuddart in #226
- Codify the GUI responsiveness rules: guidelines section, AGENTS.md invariants, processEvents ban by @BenHuddart in #227
- Mark 19 tests/gui/ files missing pytest.mark.gui by @BenHuddart in #225
Full Changelog: v0.7.0...v0.8.0