Release Notes¶
main¶
Oct. 24, 2024 don la dieu (nega at icecube.umd.edu)¶
IceTray Release v1.12.0
Update majority of tests to I3MTRandomService (#3658)
Make I3PhotoSplineService memory sharing optional. (#3646)
Clean up doxygen warnings (#3640)
replace boost’s sha256 with std::hash (#3620)
improve log_info and log_warn messages
[testing] Add RESOURCE_LOCK properties to tests with I3PhotosplineService (#3591)
Compile cleanly with -Wall -Werror (#3540)
June 18, 2024 don la dieu (nega at icecube.umd.edu)¶
IceTray Release v1.11.0
Run mypy in CI (#3480)
mypy cleanup for reconstruction projects (#3418)
Feb. 20, 2024 don la dieu (nega at icecube.umd.edu)¶
IceTray Release v1.9.2
Adding boost-python header (#3359)
Delete mailinglist files (#3357)
Add upgrade rules to ruff (#3316)
Remove all tab characters used for indentation from python files (#3300)
Dec. 14, 2023 don la dieu (nega at icecube.umd.edu)¶
IceTray Release v1.9.0
Remove Uber Header (I3.h) (#3151)
Fix E731 lambda-assignment (#3271)
fix E703 useless-semicolon (#3266)
remove unnecessary
__future__
imports (#3243)reorder initialization to avoid warning
general python cleanups
Sep. 28, 2023 don la dieu (nega at icecube.umd.edu)¶
IceTray Release v1.8.0
Add emScale to PhotonicsHitmaker (#3091)
June 22, 2023 don la dieu (nega at icecube.umd.edu)¶
IceTray Release v1.7.0
Make I3Tray.I3Tray and icecube.icetray.I3Tray the same thing. (#3102)
Remove all references to load_pybindings (#3086)
Tilt correction for infinite muon splines (#3083)
General code cleanups
Apr. 12, 2023 don la dieu (nega AT icecube.umd.edu)¶
IceTray Release v1.6.0
Time out when locking shared memory segment (#3012)
Dec. 12, 2022 don la dieu (nega AT icecube.umd.edu)¶
IceTray Release v1.5.0
Temporarily disable misbehaving tests
Guard functions that use I3PhotoSplineService (#2981)
Implement full-tilt parameterization and associated Jacobian (#2979)
Avoid calling evaluators twice if quantilEpsilon>=1.
Revert photospline memory estimation
Photonics-service - fix typo in log_fatal()
Prevent endless loop of precision warning
Refactor into FillJacobian and revert tilt zshift calculation
Reintegrate SegmentedSplineReco (#2630)
Fix some error in new sphinx roles and other minor formatting issues
Use sphinx extension extlinks to shorten/correct links
Jul. 29, 2022 don la dieu (nega AT icecube.umd.edu)¶
IceTray Release v1.4.0
Drop support for Python 2
Clean up python bindings / modules
Shared memory fixes for tests
Reflect relicensing agreements received from authors of core IceTray modules.
Fix buffer overflows and underflows
Apply spe compensation factor for hit maker
Clean up unused variables in I3PhotonicsService. Proper handling of mirror azimuth for full tables
Allow to apply effective distance for amplitude only, which is more suited for bfr ice.
Make Photospline optional (#2847)
Fix a bug when filling jacobians that was introduced when PR2500 was merged
Avoid underflow when nothing needs to be done (#2839)
Fix buffer overflow (#2834)
Improve handling of photospline memory estimate.
Prevent accumulation of shared memory segments from running tests
Clean up documentation
Feb. 21, 2022 Tianlu Yuan¶
IceTray release v1.3.0
Add a SetEfficiencies API for setting the unshadowed fraction and DOM efficiency ala clsim.
– Chris Weaver, Tianlu Yuan
Reintegrate ice-properties branch from SVN, adding support for “effective distance” treatment of ice anisotropy
– Tianlu Yuan, Dmitry Chirkin, Marcel User, Juliana Stachurska, Jakob van Santen
Dec. 20, 2019 Alex Olivas (aolivas@umd.edu)¶
Combo Release V00-00-00
Copy over the I3PhotonicsServiceCollection from a branch so that we can use the trunk of photonics-service with the upgrade and Gen2 recos.
– Michael Larson
June 20, 2018 Juan Carlos Diaz-Velez (juancarlos@icecube.wisc.edu)¶
Release V01-04-00 * Added tilt correction. To use, set tiltTableDir to point to resources/tilt
Nov 3, 2017 Juan Carlos Diaz-Velez (juancarlos@icecube.wisc.edu)¶
Release V01-03-00 * Updated tests
March 27, 2017 Juan Carlos Diaz-Velez (juancarlos@icecube.wisc.edu)¶
Release V01-02-05
Added slicing functionality and knots-vectors retrieval (+ pybindings)
– Thorsten Gluesenkamp
fixes syntax for python3
– David Schultz
Group velocity is given to fillHessian function
Fixing failing tests
– Andrii Terliuk
Cascades are forced to have length 0
No deprecated numpy code is used
Fill out both halves of the Hessian matrix in I3PhotoSplineTable::EvalHessian()
– Jakob van Santen
April 3, 2015 Meike de With (meike.de.with@desy.de)¶
Release V01-02-01
Fixes for new I3MCTree
– David Schultz
Move .fits files to I3_TESTDATA and make tests refer to that
Small code fixes to suppress warnings
Close potential memory leak
– Don la Dieu
Small code fixes to suppress warnings
Select module coordinates by I3OMGeo object instead of requiring client code to unpack coordinates itself
– Jakob van Santen
December 12, 2014 Meike de With (meike.de.with@desy.de)¶
Release V01-02-00
Make PhotoSpline radial extents an input variable, to speed up calculations for low-energy events (DOMs far away are not expected to get any light, so just return a 0 NPE expectation for these DOMs, which saves a lot of look-up/calculation time)
– Andrii Terliuk / Moriah Tobin / Jakob van Santen
November 24, 2014 Alex Olivas (olivas@icecube.umd.edu)¶
Release V01-01-02
Recognizes I3_TESTDATA now.
Make PhotoSpline Radial Extents an input variable
September 23, 2014 Meike de With (meike.de.with@desy.de)¶
Release V01-01-01
Include standard integers
Fix block quotes in rst documentation
– Don la Dieu
January 15, 2014 Meike de With (meike.de.with@desy.de)¶
Release V01-01-00
Add the ability to add hits to an already existing map
– Alex Olivas
Make I3DummyPhotonicsService do integrals
Use the same nGroup in the Jacobian matrix calculation as in the coordinate transformation
Add a test to ensure that SelectSource() and GetProbabilityQuantiles() return bit-for-bit identical results whether or not gradients were calculated
– Jakob van Santen
V00-05-00, 2013-02-21, Emanuel Jacobi (emanuel.jacobi@desy.de)¶
Raise default DOM efficiency in I3PhotonicsHitMaker to 0.99
– Nathan Whitehorn
Allow use of SelectSource() with photorec tables – this allows reconstructions (like Millipede) that use the SelectSource() interface to use photorec tables without modification
– Chris Weaver
Remove dependency on photonics and include photonics table reader support internally
Improve logging
– Kevin Meagher
V00-04-05, 2012-09-11, Emanuel Jacobi (emanuel.jacobi@desy.de)¶
Fix after removal of global GetService()
– Nathan Whitehorn
V00-04-04, 2012-08-04, Emanuel Jacobi (emanuel.jacobi@desy.de)¶
Fix bug to allow millipede calculating gradients correctly
Future proof against deprecated GetService
– Nathan Whitehorn
V00-04-03, 2012-06-27, Emanuel Jacobi (emanuel.jacobi@desy.de)¶
Use new get_photon_delays() API in Photonics 1.7 to do much faster table-based EHE simulation
Use spline table header geometry instead of input particle types to determine table type
Rationalize construction of photonics services by replacing static Create() methods with constructors (in both C++ and Python)
– Nathan Whitehorn
Allow I3PhotonicsHitMaker to accept None/NULL photonics services for cascades/tracks. Particle types without a photonics service will be ignored.
Allow I3PhotonicsHitMaker to be configured with a I3RandomService from python. The default is still to get it from the context.
Silence clang warnings
Do not propagate dark and non-in-ice particles in PhotonicsHitMaker
– Claudio Kopper
Commented out error message when tables directory is missing.
– Juan Carlos Diaz Velez
V00-04-02, 2012-02-09, Emanuel Jacobi (emanuel.jacobi@desy.de)¶
Add a minimal after-the-fact “binned mode” to prevent this from writing quite as many MC hits when doing EHE simulation.
Add a backdoor to old-hit-maker-style binned mode for truly ludicrous numbers of photons.
Pybindings for the dummy service.
Use the default fallbacks for gradient-aware SelectSource() and company.
– Nathan Whitehorn
Fix output for SelectSource() when getAmp=false and the spline call is outside the spline boundary
– Jacob Feintzeig
Allow cascade photorec tables to be queried via SelectSource(). Since SelectSource() uses the PhotonicsService energy scaling, the mean amplitude returned by SelectSource() on level-2 cascade tables will be 13% higher than when queried with GetPhotoRecInfo(). This is intentional and correct.
Clean up GetSource()/GetProbabilityDensity() Photorec facade.
– Jakob van Santen
Restore python 2.4 compatibility
– Don La Dieu
V00-04-01, 2012-01-14, Nathan Whitehorn (nwhitehorn@icecube.wisc.edu)¶
Rearrange class hierarchy so that all services (Photonics tables, splines, the dummy service) inherit from a common abstract base class.
Make I3PhotonicsServiceCommons a private interface
Add constructors for PhotonicsSource, including one that sets one up from an I3Particle
Deprecate a variety of old APIs (e.g. FillPhotonicsSource()) made obsolete by the new PhotonicsSource constructors
Move monopole light scaling from hit-maker to photonics-service
Add a new module, I3PhotonicsHitMaker, written as a replacement for hit-maker (this is a C++ port of the former Mafioso module)
Use phys-services I3RandomService->GSL bridge instead of a hand-rolled one in I3PhotoSplineService
Default Photonics service name changed from “PhotonicsService” to “I3PhotonicsService” and now uses I3_DEFAULT_NAME()
Fix spline tables with non-default indices of refraction (i.e. all of them), courtesy of Jacob Feintzeig
Add support for infinite muon spline tables (Jacob Feintzeig)
V00-03-04, 2011-08-09, Emanuel Jacobi (emanuel.jacobi@desy.de)¶
Fix a bug where asking for the gradient on a repeated SelectSource?() call would return gibberish.
– Nathan Whitehorn
Call through to the PhotonicsCommons? version of fillPhotonicsService(). This makes sure all the fields are filled properly. Without this, the various sines and cosines of the source direction would not be filled, making the photospline gradient code return gibberish.
– Nathan Whitehorn
V00-03-03, 2011-05-25, Nathan Whitehorn (nwhitehorn@icecube.wisc.edu)¶
Improve pybindings to bind a vector of PhotonicsSource objects
Reduce burn-in period in the photospline service MH sampler
Fix bugs related to photospline probability quantiles crossing table edges
V00-03-00, 2010-12-02, Jakob van Santen (vansanten@wisc.edu)¶
Bugfix: Fix inverted coordinate system in PhotoSplineService and add a parity property to PhotonicsService that indicates whether the underlying tables were generated with positive source zenith angles (in which case table azimuth points towards negative IceCube coordinate z) or negative zenith angles (in which case it points towards positive z).
Feature: Support for querying gradients of total amplitude and probability quantiles, returned in detector coordinates. The internal Photonics coordinate system may be either cylindrical or spherical. This is only implemented in PhotoSplineService, since it knows about analytic derivatives.
Feature: Support for second derivatives, also in detector coordinates.
Feature: parallel evaluation of table value and gradient in PhotoSplineService (7 evaluations for the price of 2 with a decent vectorizing compiler).
Feature: Numpy-ified pybindings for simplified testing.
Reorganization: factored pybindings out into multiple files for better maintainability.
V00-02-00, 2010-08-27, Jakob van Santen (vansanten@wisc.edu)¶
Physics tweak: Cherenkov photons per meter 0.2% larger due to wavelength-dependent index of refraction.
Physics tweak: Use M. Kowalski’s parameterization of the effective track length for EM cascades (20% larger)
Feature: Two new functions, GetTimeDelays() and GetMeanAmplitudes(), that fill arrays of timing information. This addresses a criticism from the original PhotonicsService code review.
Feature: I3PhotoSplineService, a new subclass of I3PhotonicsService that provides smooth interpolation from spline coefficient tables. This is only compiled if photospline can be found.
Feature: Support for spherical table geometries.
Feature: Complete Python bindings.