Changelog#
In progress#
Fixes:
Hist.from_columnsbuilds anIntCategoryaxis from NumPy integer arrays instead of raisingTypeError.Stacksupports/and/=with a scalar or array, like*and*=.fill_flattened()accepts plain lists, tuples, and scalars, likefill(), instead of raisingTypeError.ChunkedHist.to_hist()keeps unset category axis labels empty, so the result can be added to theHistit came from.
Version 2.12.1#
Fixes:
Raise
KeyErrorfor a missing exact category name in a list that also contains wildcards, instead of silently dropping it. #738fill_flattened()accepts pandas DataFrames and Series. #730Raise
ValueErrorwhen.expand_cats()generates duplicate keys, including from customnamecallbacks or empty groups, instead of silently overwriting a group. #728ChunkedHist.reset()and selection keep the declared categories of non-growing chunk axes, so those categories can still be filled afterwards. #742Correct weighted Poisson intervals for integer input arrays. #732
Avoid integer overflow in unweighted Poisson intervals. #741
Integrate all categories matched by wildcard patterns. #733
Resolve wildcard category selections correctly after an ellipsis. #744
Preserve rotated side-panel count ticks. #725
Performance:
import histno longer imports pandas; pandas support is registered once pandas is loaded. #730
Documentation:
Version 2.12.0#
Features:
Remove Python 3.10 support.
Support arbitrary mutable mappings for
ChunkedHistchunk storage viachunks=; existing entries are adopted, and the store is exposed as.chunks. #715Add
.expand_cats()to split all categorical axes into a dict of histograms. #717Accept objects with
_to_uhi_in the histogram constructor. #719Add hooks for working with graphed. #721
Fixes:
Documentation:
Add
llms.txtand Markdown versions of documentation pages. #720
Version 2.11.0#
Features:
Add chunked histograms (
hist.chunked.ChunkedHist). #685Support boost-histogram 1.8, require 1.7+. #707
Support Python 3.15. #708
Expose
hist.atand documentStackname-slice semantics. #695
Fixes:
Preserve histogram subclass type through quick construction. #701
Align the main axis x-range in ratio/pull plots with user-supplied axes. #700
Take the square root for the Gaussian fit sigma initial guess. #698
Simplify the pull gradient alpha and fix the SVG coordinate format. #694
Skip empty chunks and deduplicate fill bookkeeping in chunked histograms. #691
Performance:
Reuse the weight sum in profile and avoid a mutable default storage. #693
Typing:
Annotate ndarray with generic args in chunked histograms and add a Python 3.10 mypy gate. #697
Tests/CI:
Regenerate matplotlib baselines for matplotlib 3.11. #702
Version 2.10.1#
Fixes:
Version 2.10.0#
Features:
Remove Python 3.9 support. #646
Support boost-histogram 1.7 (including
MultiCell) and drop boost-histogram 1.5 #667
Typing:
Version 2.9.2#
Fixes:
Tests/CI:
Version 2.9.1#
Fixes:
Improve error message for invalid weighted fill #654
Improved error message for missing axes in
Hist.fill#651Misaligned ratio plots for
Intaxes #655
Tests:
CI:
GHA release notes now require bot suffix #658
Avoid deprecation warning from pyparsing in some jobs #657
Documentation:
Version 2.9.0#
Features:
Support for boost-histogram 1.6, drop support for Python 3.8, boost-histogram < 1.5, add support for Python 3.14/3.14t #597 #618, #631
Support serialization with boost-histogram 1.6 #627
Add
legend=Truefor plot1d stacked #622
Fixes:
Handle fill-flattened for non-user args #612
Handle string args for fill-flattened #629
Label/name propagation fixed for casting histograms #630
Tests:
CI:
Version 2.8.1#
Features:
Allow wildcards in StrCat list indexing by @andrzejnovak #601
Tests:
Limit dask version #596
Version 2.8.0#
Features:
Support boost-histogram 1.5; drop support for Python 3.7, add support for Python 3.13 / 3.13t #582
Fixes:
Support IntCat for plot1d overlay #419
Version 2.7.3#
This release fixes an issue with Python 3.12 pulling in NumPy 2 pre-releases.
Fixes:
Docs:
Clarify sample term required for Mean storage #536
CI:
Version 2.7.2#
Support boost-histogram 1.4.0 in addition to 1.3.x, including Python 3.12, flow disabling for categories, and integer arrays required for integer axes #535 and #532
Add a
.Tshortcut #521Support ND quick constructs being mistakenly passed in #528
Add
boostnox job to test upstream changes easily #533
Version 2.7.1#
Version 2.7.0#
Features:
Various other items:
Version 2.6.3#
Version 2.6.2#
Version 2.6.1#
Version 2.6.0#
Using boost-histogram 1.3
Fix runtime dependency on matplotlib when not plotting #353
Fix
.plotshortcut failure #368New nox sessions:
regenerateandpylintUpdate tests for latest matplotlib
Version 2.5.2#
Version 2.5.1#
Version 2.5.0#
Stacks support axes, math operations, projection, setting items, and iter/dict construction. They also support histogram titles in legends. Added histoprint support for Stacks. #291 #315 #317 #318
Added
name=andlabel=to histograms, include Hist arguments in QuickConstruct. #297AxesTuple now supports bulk name setting,
h.axes.name = ("a", "b", ...). #288Added
hist.newalias forhist.Hist.new. #296Added
"efficiency"uncertainty_typeoption forratio_plotAPI. #266 #278
Smaller features or fixes:
Dropped Python 3.6 support. #194
Uses boost-histogram 1.2.x series, includes all features and fixes, and Python 3.10 support.
No longer require scipy or iminuit unless actually needed. #316
Improve and clarify treatment of confidence intervals in
intervalssubmodule. #281Use NumPy 1.21 for static typing. #285
Support running tests without plotting requirements. #321
Version 2.4.0#
Support selection lists (experimental with boost-histogram 1.1.0). #255
Support full names for QuickConstruct, and support mistaken usage in constructor. #256
Add
.sort(axis)for quickly sorting a categorical axis. #243
Smaller features or fixes:
Support nox for easier contributor setup. #228
Better name axis error. #232
Fix for issue plotting size 0 axes. #238
Fix issues with repr information missing. #241
Fix issues with wrong plot shortcut being triggered by Integer axes. #247
Warn and better error if overlapping keyword used as axis name. #250
Along with lots of smaller docs updates.
Version 2.3.0#
Add
plot_ratioto the public API, which allows for making ratio plots between the histogram and either another histogram or a callable. #161Add
.profileto compute a (N-1)D profile histogram. #160Support plot1d / plot on Histograms with a categorical axis. #174
Add frequentist coverage interval support in the
intervalsmodule. #176Allow
plot_pullto take a more generic callable or a string as a fitting function. Introduce an option to perform a likelihood fit. Write fit parameters’ values and uncertainties in the legend. #149Add
fit_fmt=toplot_pullto control display of fit params. #168Support
<prefix>_kwarguments for setting each axis in plotting. #193Cleaner IPython completion for Python 3.7+. #179
Version 2.2.1#
Version 2.2.0#
Version 2.1.1#
Fix density (and density based previews) #134
Version 2.1.0#
Support shortcuts for setting storages by string or position #129
Updated dependencies:
boost-histogram0.11.0 to 0.13.0.Major new features, including PlottableProtocol
histoprint>=1.4 to >=1.6.mplhep>=0.2.16 when[plot]given
Version 2.0.1#
Version 2.0.0#
First release of Hist.