Release notes#

Version 0.19.1 (2026-01-20)#

Emergency patch removing the minimum constraint on the scale parameter of the exponential models, in order to be able to fit rising exponentials.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/92

Version 0.19.0 (2026-01-19)#

Added the fwhm() method to the Histogram1d class to calculate the full width at half maximum of a one-dimensional histogram.
Cumulative distribution function and percent point function calculation implemented for Histogram1d.
Minimum coverage interval calculation implemented for Histogram1d.
Pull requests merged and issues closed:

Version 0.18.0 (2025-12-20)#

Changed parametrization of the PowerLaw model to include an arbitrary pivot that can be set at creation time.
Probit fitting model added.
Added the Histogram3d class for three-dimensional histograms.
New slice1d(), project_mean(), and project_statistics() methods added in the histogram base class.
Minor warning fixes in various places (docs and unit tests).
Updated copyright notice.
Added team page in the documentation.
Pull requests merged and issues closed:

Version 0.17.2 (2025-12-11)#

Fix for issue #74 (error when calling ufloat with no errors).
Correlated fit parameters added to the FitStatus object.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/76
- https://github.com/lucabaldini/aptapy/issues/74

Version 0.17.1 (2025-12-09)#

Moved random_histogram() to the AbstractFitModelBase class
Signature updated to require histogram edges instead of the number of bins.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/75
- https://github.com/lucabaldini/aptapy/issues/67

Version 0.17.0 (2025-12-05)#

Refactored GaussianForestBase parameter structure from individual amplitudes to a shared amplitude with intensity ratios.
Moved GaussianForestBase from models.py to modeling.py.
Significant refactoring to allow for freezing parameters in line forests.
Added the fit_iterative() method to the GaussianForestBase class.
Added the fwhm() method to the Gaussian model class.
Modified the fwhm() method of the GaussianForestBase to return the real FWHM and not the resolution of the line.
Added unit tests.
Pull requests merged and issues closed:

Version 0.16.0 (2025-12-03)#

Added the GaussianForestBase and Fe55Forest model classes.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/62
- https://github.com/lucabaldini/aptapy/issues/47

Version 0.15.1 (2025-11-21)#

Fix for a bug in the gaussian iterative fitting in a range.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/issues/60

Version 0.15.0 (2025-11-21)#

Naming order for polynomial coefficients changed to match standard mathematical notation (e.g., c2 for quadratic term, c1 for linear term, c0 for constant term).
Keeping track of the covariance matrix in the FitResult object.
Added numerical Jacobian calculation in the base class for fit models.
Added capability to plot confidence bands around fit models.
Added fit_iterative() method to the Gaussian model class to perform iterative fitting.
Added the set_content() method to AbstractHistogram to fill a histogram with binned data
Slight change to the main fitting interface, and fit_histogram() function removed.
Added the from_amptek_file() method to create a Histogram1d from an MCA8000A multichannel analyzer output file.
Pull requests merged and issues closed:

Version 0.14.0 (2025-11-14)#

Major refactoring of the modeling infrastructure, introducing a more structured and extensible framework for fitting models.
New abstract base classes (AbstractSigmoidFitModel, AbstractCRVFitModel) to support different model families
Added 40+ new model classes wrapping scipy.stats continuous random variables
Refactored existing models to use a consistent primitive() method instead of integral()
Reorganized tests into category-specific files (simple, sigmoid, scipy models)
Updated parameter naming.
Pull requests merged and issues closed:

Version 0.13.0 (2025-11-01)#

Major refactoring of the modeling module (now split into the models and modeling modules, the former containing all the core functionality and latter the actual models) to improve code organization and maintainability.
Erf and ErfInverse models have been renamed.
New models from the exponential family, including the StretchedExponential and its complement, and all exponential models now supporting a generic origin.
Added facility to draw random samples from fit models, in order to facilitate model validation and testing.
Base class AbstractPlottable now correctly propagated to all fit models.
Detailed mathematical formulae added to the docstrings of all model classes.
Fix for issue #33 (cannot freeze parameters for composite models).
Components within composite models now accessible via indexing (e.g., composite_model[0] returns the first component).
Added a unit test for composite model parameter freezing.
New plot_components argument in the plot() method of fit models to control individual components of composite models.
Arguments names for the integral() methods of fit models updated from xmin, xmax to x1, x2 for consistency.
Enhanced unit test coverage for the modeling module.
Pull requests merged and issues closed:

Version 0.12.0 (2025-10-29)#

Added support for subtracting callable models from Histogram1d objects (this is useful, e.g., to create a residual histogram wrt a fit model).
Introduced subplot_vstack() and residual_axes() functions for creating multi-panel plots.
Enhanced the fit() method to support excluding intervals when xmin > xmax.
Improved histogram copying functionality to allow optional relabeling.
Updated documentation and example gallery.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/32
- https://github.com/lucabaldini/aptapy/issues/30

Version 0.11.0 (2025-10-28)#

New mechanism for matplotlib style management in the plotting module.
Replace the configure() function with apply_stylesheet() and stylesheet_context() context manager for better control over style application.
Shipping a custom dark theme and bundling the Humor Sans font for xkcd-style stylesheet.
Dependencies on cycler and loguru packages removed.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/31
- https://github.com/lucabaldini/aptapy/issues/27

Version 0.10.2 (2025-10-27)#

Default vertical alignment for the text in the ConstrainedTextMarker class changed from “center” to “bottom” to avoid overlapping with the underlying trajectory for nearly horizontal data sets.
Fixed a bug in the interactive cursor logic triggered by mouse events outside the axes area.
Fixed random seed in tests/test_hist.py to ensure consistent results across runs.
Added a –interactive option to pytest to allow keeping matplotlib figures open after test execution for interactive inspection.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/29

Version 0.10.1 (2025-10-27)#

Small refactoring in modeling.py.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/28

Version 0.10.0 (2025-10-27)#

New AbstractPlottable base class with standard plot() and abstract _render() methods.
Refactored AbstractFitModelBase, AbstractHistogram, and StripChart, as well as all fit models, to inherit from AbstractPlottable.
Simplified plotting range management for fit models.
Example gallery revamped.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/26

Version 0.9.3 (2025-10-23)#

Added dependencies on sphinxcontrib-programoutput and nox.
Added new sections in the documentation for the installation and development workflows.
Refactored nox tasks for better build cleanup functionality
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/25

Version 0.9.2 (2025-10-22)#

Added error handling in ConstrainedTextMarker.move() to gracefully hide markers when trajectory calculations fail (e.g., when extrapolating outside data range).
Enhanced StripChart.spline() to support configurable extrapolation behavior via the ext parameter.
Refactored last_line_color() to accept an optional axes parameter, improving reusability and eliminating redundant plt.gca() calls.
Updated unit tests.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/24

Version 0.9.1 (2025-10-21)#

Fixed package logo not appearing on PyPI by using absolute URL in README.md.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/22
- https://github.com/lucabaldini/aptapy/issues/21

Version 0.8.0 (2025-10-20)#

Public interface for the StripChart class improved: append() and extend() merged into put(), that should handle both single values and iterables.
Added __len__() method to support len() on StripChart objects.
Comprehensive test coverage for various input types and error conditions.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/20
- https://github.com/lucabaldini/aptapy/issues/19

Version 0.7.1 (2025-10-20)#

Fix for issue #15 (traceback when plotting empty histograms).
set_max_length() method added to strip charts to allow changing the max length of the underlying deques.
Avoid catching bare exception in __init__.py.
Pull requests merged and issues closed:

Version 0.7.0 (2025-10-17)#

Strip chart formatting on the x-axis improved, and full refactoring of the StripChart class, with the addition of the EpochStripChart subclass.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/14
- https://github.com/lucabaldini/aptapy/issues/13

Version 0.6.0 (2025-10-17)#

Addition of VerticalCursor and ConstrainedTextMarker classes for interactive plotting.
Enhancement of StripChart with method chaining and spline interpolation capabilities.
Comprehensive test coverage for the new cursor functionality.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/12

Version 0.5.0 (2025-10-12)#

Added init_parameters method to most model classes.
Updated import structure to use scipy.special module directly instead of importing erf.
Added comprehensive test coverage for the new parameter initialization functionality.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/10
- https://github.com/lucabaldini/aptapy/issues/9

Version 0.4.0 (2025-10-11)#

Added 2-dimensional histogram example.
Adds several new model classes (Quadratic, PowerLaw, Exponential, Erf, ErfInverse).
Implements analytical integration methods for models where possible, with a fallback to numerical integration in the base class.
Updates the FitStatus class with a completion check method.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/7

Version 0.3.2 (2025-10-09)#

Adding binned_statistics method in AbstractHistogram base class to calculate statistics from histogram bins
Adds extensive test coverage in both 1D and 2D histogram test functions with statistical validation
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/6

Version 0.3.1 (2025-10-09)#

Minor changes.

Version 0.3.0 (2025-10-08)#

New strip-chart facilities added.
Introduction of model summation capability through operator overloading
Refactored class hierarchy with new abstract base classes
Enhanced parameter compatibility checking methods
Improved histogram integration for fitting
Adds sphinx-gallery integration with 5 example scripts demonstrating histogram and fitting functionality
Improves statistical analysis by adding p-value calculations and fixing degrees of freedom calculations
Updates test assertions to include p-value validation
Pull requests merged and issues closed:

Version 0.2.0 (2025-10-06)#

New histogram facilities added.
Pull requests merged and issues closed:
- https://github.com/lucabaldini/aptapy/pull/2

Version 0.1.1 (2025-10-03)#

Initial release on PyPI.