Changelog¶
0.1.6 (2026-11-04)¶
Changed (breaking)
DataStream.compute_statistics()now always computes the standard error of the mean assd(block_means) / sqrt(ess_blocks), whereess_blocksis the Geyer effective sample size of the block means. In 0.1.5 it divided by the raw number of blocks whenever the window came from the Ljung-Box search, whether or not the test passed. That test has low power for typical block counts and often passes while the block means are still correlated, so the standard error could be too small.mean_uncertaintyand the confidence intervals are therefore larger for almost every call. The means do not change.The default block window is now
w = round(tau_int)(the “tau-window” rule) instead of the Ljung-Box search. This changeswindow_size,n_short_averagesandess_blocks.independence_statusis now"tau_window". The Ljung-Box results are still reported, but no longer affect the window or the standard error.se_methodis now always"ess_blocks"."iid_blocks","iid_blocks_best_p"and"ess_blocks_fallback"are no longer returned.Ensemble statistics follow the same approach:
Pooled block means (T1) always use the ESS-corrected standard error (
se_method="sem_ess", previously"sem_n"when all members passed Ljung-Box).The ensemble average (T0) and inverse-variance weighted member means (T2) by default use a block window based on a correlation time pooled over all members (
tau_mode="pooled").
MeanVariationTrimStrategynow uses the same defaults asRobustWorkflow:decor_multiplier=6.0(was 2.0),fudge_fac=0.1(was 1e-6) andfinal_smoothing_window=10(was 5). This changes the detected SSS start when the strategy is created without these arguments, includingbuild_trim_strategy("mean_variation")andDataStream.trim(method="mean_variation").RobustWorkflowresults do not change.The regression baselines in
tests/*/expected/were regenerated.
Added
DataStream.compute_statistics(window_rule=...):"tau"(default) or"autotune"for the legacy Ljung-Box window search. Both rules use the ESS-based standard error. An explicitwindow_sizeoverrides the rule.Ensemble statistics:
tau_mode="pooled"|"per_member"for T0 and T2 (use"per_member"when members really differ, e.g. parameter scans), andpooled_ess_method="concat"|"sum"for T1. T0 metadata also reportstau_modeandtau_star.quends.base.utils.tau_window_blocks()andquends.base.utils.pooled_tau_int()(pooled correlation time for homogeneous ensembles).quends.MEAN_VARIATION_DEFAULTS: read-only mapping of the start-of-SSS detection defaults, shared byMeanVariationTrimStrategyandRobustWorkflow.
Fixed
Tau-window results no longer get the misleading warning “Too few blocks for independence test; SE via Geyer ESS.”.
_process_column()now rejects an unknownwindow_ruleeven when an explicit window is given.
Documentation
New README section “Detecting the Start of Statistical Steady State”: when to use
RobustWorkflow, how themean_variationheuristic works, theverbosity=2diagnostic plots, and how to tune its hyperparameters.The
DataStream.compute_statistics(),Ensemble, ensemble statistics and ensemble statistics workflow docstrings now describe the tau-window blocking and the ESS-based standard error.
Security
Added
pillow>=13.0.0as a dependency.Updated locked dependencies to patched releases: Black 26.10.1, certifi 2026.10.23, filelock 4.2.7, Pillow 13.0.0, Tornado 6.6, and virtualenv 21.15.2.
0.1.5 (2026-10-15)¶
Changed (breaking)
The minimum supported Python version is now 3.12 (previously 3.8). Python 3.8 through 3.11 are no longer supported.
Development
Updated the dependency lockfile for Python 3.12 and later.
Updated the test and documentation CI workflows to run on Python 3.12.
0.1.4 (2026-10-08)¶
Removed (breaking)
The
max_lag_fracandautocorr_sig_levelarguments were removed fromRobustWorkflowandMeanVariationTrimStrategy. They no longer had any effect on the decorrelation time, which is computed byDataStream.compute_decorrelation_time(). Passing either argument now raises aTypeError. Remove them from existing calls.
Changed
In
MeanVariationTrimStrategy, the smoothing window used to detect the start of statistical steady state is nowdecor_multiplier * tau_int, capped at the signal length (and at least 3 points). Before, it was also capped atmax_lag_frac * n_pts. With the defaultdecor_multiplierthis can give a wider smoothing window, and so a different (often later) SSS start time and different statistics than in 0.1.3.With
verbosity > 1, the autocorrelation plot now covers1.5 * tau_intlags (capped at the signal length), with a dashed vertical line at the decorrelation time and a legend.
Documentation
Added a full parameter description to the
MeanVariationTrimStrategydocstring, and fixed thetrim()parameter docs.Fixed the documented default of
smoothing_window_correctioninRobustWorkflow(0.5, not 0.8) and clarifieddecor_multiplier.Removed the obsolete arguments from the tutorial notebooks and the
robustworkflow_advanced_guide.pyexample script.
0.1.3 (2026-10-07)¶
Changed
The integrated autocorrelation time (
tau_int) estimate now extends the autocorrelation lag window when needed. The ACF is first computed up tomin(n // 4, 2000)lags. If the Geyer positive-pair stopping criterion is not reached, the number of lags is doubled repeatedly, up to the full series length (n - 1).As a result,
tau_intcan be larger, and the effective sample size smaller, than in 0.1.2 for strongly correlated series whose initial lag window was too short.
Fixed
RobustWorkflow.process_data_stream()no longer raisesIndexError: single positional indexer is out-of-boundswhenstart_timeis later than the last valid data point. In that case it now returns NaN statistics ifoperate_safe=True. Ifoperate_safe=False, it returns an ad-hoc estimate over the tail of the data stream, set byno_sss_tail_fraction. Both cases report the new status"StartTimeBeyondData".
Warnings
A
UserWarningsaying thattau_intis potentially under-estimated is now raised only when the Geyer criterion is not reached even with the full-length ACF. The same warning appears in thecompute_statistics()result metadata.This replaces the previous “decorrelation time is large compared to the max lag” warning, which was based on a
tau_int >= 0.5 * nlagsrule.
Documentation
Updated the README section on publishing to PyPI to describe the automated GitHub Release workflow.
Synchronized
__version__insrc/quends/__init__.pywith the version inpyproject.toml.
0.1.2¶
0.1.1¶
0.1.0 (2025-02-28)¶
First release on PyPI.