Release notes#

Version 0.17.1 (2026-05-26)#

Version 0.17.0 (2026-05-18)#

  • New position reconstruction algorithm based on the likelihood of the charge diffusion model implemented.

  • New module likelihood.py containing the code for the likelihood evaluation added.

  • Unified position reconstruction algorithms calibration file and the command into calibgen position.

  • New clustering algorithm ClusteringHex implemented. This clustering always contains seven pixels: the seed pixel and all its six neighbors.

  • New module position.py containing the core implementation of the position reconstruction algorithms added. As a consequence, the eta.py module has been deleted.

  • Pull requests merged and issues closed:

Version 0.16.3 (2026-05-14)#

  • Refactored zero suppression threshold to be interpreted as number of standard deviations of the noise instead of a fixed number.

  • Added check of the calibration matrices at the readout initialization to fill the empty pixels, in order to avoid problems in the reconstruction.

  • New algorithm added to calibrate the pixel equalization factors. This algorithm calculates the gain equalization factors without any assumption on the source spectrum, by analyzing the energy distributions of one-pixel events and comparing them across the pixels.

  • calibgen equalization now accepts the argument algorithm to choose between two different calibration algorithms: “relative”, without using the source spectrum, and “absolute”, that needs the source spectrum PDF to be passed as an argument.

  • Pull requests merged and issues closed:

Version 0.16.2 (2026-05-12)#

  • Package dependencies updated: joblib and iminuit added.

  • New calibration matrix type equalization added in caldb.py to store the pixel equalization factors.

  • New CalibrateEqualization class to calibrate the pixel equalization factors. This is performed with a likelihood fit using the source spectrum probability density function (PDF) and the observed energy of the events. Corresponding CLI command calibgen equalization added to generate equalization calibration files from event files.

  • The CalibrateGain class has been refactored. Now it produces a gain calibration matrix from an equalization calibration matrix, by using the ADC to eV conversion and the chip sensor material. The gain matrix is expressed in ADC counts per electron. Corresponding CLI command calibgen gain refactored to reflect this change.

  • Some CLI commands refactored to use the equalization calibration file instead of the gain calibration file (reconstruct, display, calibgen eta).

  • New pdf.py module containing facilities to calculate the PDF of the source spectrum and store it in a .npz file.

  • Added calibspec CLI command to create a source spectrum PDF file from a reconstructed event file. The PDF is saved as a .npz file.

  • New adc method to calculate the total ADC counts of the event added to ReconEvent. Now a reconstructed file contains both the ADC counts and the energy in eV of the events (if the conversion factor from ADC to eV is available in the equalization calibration file).

  • New energy method to calculate the total energy of the event added to Cluster. The conversion factor from ADC to eV is passed by ClusteringNN.

  • Pull requests merged and issues closed:

Version 0.16.1 (2026-05-12)#

  • New stats.py module with some common statistical functions, most notably a class to accumulate the running statistics using the Welford algorithm.

  • calibgen dark now accepts the argument algorithm to choose between two different algorithms to calibrate the noise and pedestal of the readout chip: “welford”, to use Welford’s online algorithm, that updates the mean and variance of the counts distribution on the fly, and “fit”, that updates a 3D histogram of the counts distribution and fits it with a Gaussian to extract the noise and pedestal.

  • Pull requests merged and issues closed:

Version 0.16.0 (2026-05-05)#

Version 0.15.0 (2026-04-23)#

Version 0.14.0 (2026-03-03)#

  • Added classes CalibrationMatrixGain and CalibrationMatrixNoise to use gain and noise matrices to support non-uniform pixel response in the readout chip.

  • Command calibrate added to the CLI interface to use a DigiFile to calibrate the gain response and noise of the readout chip.

  • Implemented the possibility to use the gain calibration matrix to simulate and reconstruct data with non-uniform pixel response, with the option –map_gain_file in the cli interface.

  • ClusteringBase modified to accept a HexagonalReadoutBase instance instead of HexagonalGrid. This guarantees that the reconstruction can be performed even with non-uniform pixel gain.

  • Cluster now has col and row arguments in the constructor, in order always carry the info about the logical coordinates of the event.

  • New legacy.py module containing the code to convert .mdat3 files to .h5 files.

  • New command convert to convert .mdat3 files to .h5 files using the CLI.

  • Added an actual interactive event display with some GUI.

  • Added support for random access in event files.

  • Small cleanup in the fileio.

  • Pull requests merged and issues closed:

Version 0.13.3 (2026-02-10)#

  • Added rough support for DigiEventCircular objects in the event display.

  • New resolution.py and calibration.py modules incorporating some common functions that were previously living in the script area.

  • New script to calculate the EEF for events reconstructed with the different algorithms, disaggregated by cluster size.

  • –max_neighbors option added to the cli interface.

  • eta reconstructions modified to use the centroid for events with more than three pixels.

  • Two parameters added to the eta reconstruction to control the transition from linear to probit reconstruction for two and three pixel events.

  • New suppression step added to accept at most three pixels in a cluster, with the condition that all the pixels are neighbors.

  • Pull requests merged and issues closed:

Version 0.13.2 (2026-02-04)#

Version 0.13.1 (2026-01-30)#

  • Added scripts to calibrate the eta function.

  • Added more command-line options to the cli interface to control the reconstruction with the eta function.

  • Bug fix in the hexagonal beam implementation.

  • Pull requests merged and issues closed:

Version 0.13.0 (2026-01-28)#

  • New command-line interface—this adds the new cli and tasks modules, and removes the old app module, as well as all the application scripts in the bin folder. This is the biggest single change in this release, and has a number of ramifications, with many files and modules affected.

  • New base module, containing a few base classes and utilities for general use, including a TypeProxy class to handle dynamic typing.

  • Major refactoring of the classes in the readout module, mainly for making the signature of the read() method consistent across all types.

  • Refactoring of the source and sensor classes.

  • Sparse readout and all associated data structures (including the event and file io machinery) removed, as deemed un-necessary.

  • pprint module renamed as pretty to avoid confusion with the standard library pprint module.

  • Limiting the pytest target in the ci workflow.

  • Documentation and unit tests updated.

  • General cleanup and linting, with removal of obsolete files such as test_optimize_sim.

  • Pull requests merged and issues closed:

Version 0.12.0 (2025-12-20)#

Version 0.11.0 (2025-11-19)#

  • Major repository restructuring to modernize the package layout and tooling infrastructure.

  • Adoption of the “src” layout for the package.

  • Migrate from a legacy setuptools-based structure to a modern pyproject.toml-based build system.

  • Introduced nox for task automation (testing, linting, documentation building).

  • Rewrote release tooling in tools/release.py with improved version management.

  • Added pytest fixtures in tests/conftest.py for test data handling and matplotlib figure management.

  • Consolidated logging configuration into src/hexsample/logging_.py module.

  • Old setup files removed in favor of an editable install via pip.

  • Old Makefile removed.

  • Documentation sphinx theme changed.

  • __init__.py file cleaned up.

  • Full linting with ruff and pylint.

  • github workflows updated.

  • Pull requests merged and issues closed:

Version 0.10.0 (2025-11-13)#

  • Adding aptapy as a new dependency in both requirements.txt and pyproject.toml.

  • Removing internal modules: hexsample/hist.py, hexsample/modeling.py, hexsample/plot.py and their test files.

  • Updating imports across all test files, source files, and command-line scripts to use aptapy.hist, aptapy.models, and aptapy.plotting.

  • Modified API calls to match aptapy’s interface.

  • Updates CI/CD configuration to use Ubuntu 22.04 and test on Python 3.7 and 3.13.

  • Pull requests merged and issues closed:

Version 0.9.0 (2025-04-30)#

Version 0.8.1 (2023-12-12)#

Version 0.8.0 (2023-12-07)#

Version 0.7.0 (2023-10-25)#

  • Major refactoring of the modeling framework, with no (intentional) modification to the public API.

  • New FitStatus class, refactoring of the FitModelBase class, with fit() and fit_histogram() now class members (as opposed to loose functions in the fitting module).

  • Fit with parameter bounds now supported.

  • Specific class for a double gaussian fit, with sensible initial values.

  • Pull requests merged and issues closed:

Version 0.6.0 (2023-10-19)#

Version 0.5.2 (2023-10-18)#

  • New attempt at compiling the docs on github pages.

Version 0.5.1 (2023-10-18)#

Version 0.5.0 (2023-10-17)#

Version 0.4.0 (2023-10-16)#

  • This is a major rework of the sampling, trigger and digitization facilities allowing for a simulation speedup of almost an order of magnitude, without loss of performance.

  • Digitization machinery refactored in order to avoid working with large sparse arrays (in pixel and minicluster space) full of zeroes.

  • Generation of the noise moved at the end of the digitization process.

  • Hexagonal sampling largely rewritten to avoid the use of numpy.histogram2d.

  • Trigger machinery reworked to accommodate the previous changes.

  • Comparison operator defined for Padding, RegionOfInterest and DigiEvent in order to be able to make strict comparisons between output digi files.

  • Seed for a small utility to compare digi files added.

  • Pull requests merged and issues closed:

Version 0.3.2 (2023-10-16)#

Version 0.3.1 (2023-10-13)#

Version 0.3.0 (2023-10-13)#

Version 0.2.0 (2023-10-12)#

Version 0.1.0 (2023-10-10)#