Skip to content

Latest commit

 

History

History
357 lines (330 loc) · 22.1 KB

CHANGELOG.md

File metadata and controls

357 lines (330 loc) · 22.1 KB
Raw
  • 24.x.x
  • Bug fixes:
    • Fix bug with 'median' and 'mean' methods in Masker averaging over the wrong axes.
    • SPDHG gamma parameter is now applied correctly so that the product of the dual and primal step sizes remains constant as gamma varies (#1644)
    • Allow MaskGenerator to be run on DataContainers (#2001)
    • Make Paganin Processor work with AcquistionData with one angle (#1920)
    • Fix bug passing kwargs to PDHG (#2010)
  • Enhancements:
    • Removed multiple exits from numba implementation of KullbackLeibler divergence (#1901)
    • Updated the SPDHG algorithm to take a stochastic Sampler(#1644)
    • Updated the SPDHG algorithm to include setters for step sizes (#1644)
  • Dependencies:
    • Added scikit-image to CIL-Demos conda install command as needed for new Callbacks notebook.
  • Changes that break backwards compatibility:
    • Deprecated norms and prob in the SPDHG algorithm to be set in the BlockOperator and Sampler respectively (#1644)
    • The run method in the cil algorithm class will no longer run if a number of iterations is not passed (#1940)
    • Paganin processor now requires the CIL data order (#1920)

  • 24.2.0

    • New Features:
      • Added SVRG and LSVRG stochastic functions (#1625)
      • Added SAG and SAGA stochastic functions (#1624)
      • Allow SumFunction with 1 item (#1857)
      • Added PD3O algorithm (#1834)
      • Added Barzilai-Borwein step size rule to work with GD, ISTA, FISTA (#1859)
      • Added callback optimisation.utilities.callbacks.EarlyStoppingObjectiveValue which stops iterations if an algorithm objective changes less than a provided threshold (#1892)
      • Added callback optimisation.utilities.callbacks.CGLSEarlyStopping which replicates the automatic behaviour of CGLS in CIL versions <=24. (#1892)
      • Added labels module with ImageDimension, AcquisitionDimension, AcquisitionType, AngleUnit, FillType (#1692)
    • Enhancements:
      • Use ravel instead of flat in KullbackLeibler numba backend (#1874)
      • Upgrade Python wrapper (#1873, #1875)
      • Updated the documentation for the algorithm base class (#1809)
      • Add checks on out argument passed to processors to ensure correct dtype and size (#1805)
      • Internal refactor: Replaced string-based label checks with enum-based checks for improved type safety and consistency (#1692)
      • Internal refactor: Separate framework into multiple files (#1692)
      • Allow the SIRT algorithm to take initial=None (#1906)
      • Add checks on equality method of AcquisitionData and ImageData for equality of data type and geometry (#1919)
      • Add check on equality method of AcquisitionGeometry for equality of dimension labels (#1919)
    • Testing:
      • New unit tests for operators and functions to check for in place errors and the behaviour of out (#1805)
      • Updates in SPDHG vs PDHG unit test to reduce test time and adjustments to parameters (#1898)
      • Drop Jenkins in favour of GHA for conda builds (#1914)
      • New unit tests for DataContainer, AcquisitionData and ImageData to check equality method (__eq__) behaves as expected (#1919)
    • Bug fixes:
      • ImageData removes dimensions of size 1 from the input array. This fixes an issue where single slice reconstructions from 3D data would fail due to shape mismatches (#1885)
      • Make Binner accept accelerated=False (#1887)
      • Added checks on memory allocations within FiniteDifferenceLibrary.cpp and verified the status of the return in GradientOperator (#1929)
      • Build release version of cilacc.dll for Windows. Previously was defaulting to the debug build (#1928)
      • Armijo step size rule now by default initialises the search for a step size from the previously calculated step size (#1934)
    • Changes that break backwards compatibility:
      • CGLS will no longer automatically stop iterations once a default tolerance is reached. The option to pass tolerance will be deprecated to be replaced by optimisation.utilities.callbacks (#1892)

  • 24.1.0

    • New Features:
      • Added method to plot filter in GenericFilteredBackProjection (#1667)
      • Added wavelet operator, wrapping PyWavelets operator as a CIL operator (#1615)
      • Added PaganinProcessor processor, to perform phase retrieval from phase contrast images (#1737)
      • Added L1Sparsity function, allowing calculations of |Ax-b|_1 and its proximal, in the case of orthogonal operators, A (#1618)
      • Options in algorithms GD, ISTA and FISTA to pass a cil.optimisation.utilities.StepSizeRule or a cil.optimisation.utilities.Preconditioner(#1768)
      • an implementation of the Armijo Rule as a child class of cil.optimisation.utilities.StepSizeRule (#1768)
      • Sensitivity preconditioners added as child classes of cil.optimisation.utilities.Preconditioner(#1768)
    • Enhancements:
      • Added geometry property to BlockDataContainer. Adds __eq__ to BlockGeometry (#1799)
      • Raises error in BlockDataContainer.pnorm if the shape of the containers is not the same (#1799)
      • Operators and functions now also return when out is specified (#1742)
      • The CIL function class now has a __neg__ function, so you can write -YourFunction(x) rather than -1*YourFunction(x) (#1808)
      • Added documentation for the Partitioner to framework.rst (#1828)
      • Added CIL vs SIRF tests comparing preconditioned ISTA in CIL and MLEM in SIRF (#1823)
      • Update to CCPi-Regularisation toolkit v24.0.1 (#1868)
    • Bug fixes:
      • gradient descent update_objective called twice on the initial point.(#1789)
      • ProjectionMap operator bug fix in adjoint and added documentation (#1743)
      • BlockOperator that would return a BlockDataContainer of shape (1,1) now returns the appropriate DataContainer. BlockDataContainer direct and adjoint methods accept DataContainer as parameter (#1802).
      • BlurringOperator: remove check for geometry class (old SIRF integration bug) (#1807)
      • The ZeroFunction and ConstantFunction now have a Lipschitz constant of 1. (#1768)
      • Update dataexample remote data download to work with windows and use zenodo_get for data download (#1774)
    • Changes that break backwards compatibility:
      • Merged the files BlockGeometry.py and BlockDataContainer.py in framework to one file block.py. Please use from cil.framework import BlockGeometry, BlockDataContainer as before (#1799)
      • Bug fix in FGP_TV function to set the default behaviour not to enforce non-negativity (#1826).

  • 24.0.0

    • Update to new CCPi-Regularisation toolkit v24.0.0. This is a backward incompatible release of the toolkit.
    • CIL plugin support for TIGRE version v2.6
    • CIL plugin support for ASTRA-TOOLBOX version v2.1
    • Dropped support for python 3.8 and 3.9
    • Added support for python 3.11 and 3.12
    • Dropped support for numpy 1.21 and 1.22
    • Added support for numpy 1.25 and 1.26
    • Set CMake Policy CMP0148 to OLD to avoid warnings in CMake 3.27
    • AcquisitionGeometry prints the first and last 10 angles, or all if there are 30 or less, rather than the first 20
    • Added a weight argument to the L1Norm function
    • Allow reduction methods on the DataContainer class to accept axis argument as string which matches values in dimension_labels
    • Added the functions set_norms and get_norms to the BlockOperator class
    • Internal variable name change in BlockOperator to aid understanding
    • Fixed formatting errors in the L2NormSquared and LeastSquares documentation
    • Bug fix for BlockDataContainer as iterator
    • Dropped support for IPP versions older than 2021.10 due to header changes
    • Fix build include directories
    • Proximal of MixedL21Norm with numpy backend now accepts numpy ndarray, DataContainer and float as tau parameter
    • ZeroOperator no longer relies on the default of allocate
    • Bug fix in SIRF TotalVariation unit tests with warm_start
    • Allow show2D to be used with 3D DataContainer instances
    • Added the a Sampler class as a CIL optimisation utility
    • Update documentation for symmetrised gradient
    • Added documentation for CompositionOperator and SumOperator
    • Updated FISTA and ISTA algorithms to allow input functions to be None
    • Bug fix in the adjoint of the Diagonal Operator for complex values
    • Update conda build action to v2 for 2.5x quicker builds
    • Add docker image, test & push to ghcr.io/tomographicimaging/cil
    • Quality metrics have added mask option
    • Bug fix for norm of CompositionOperator
    • Functions updated to use sapyb
    • Fix KullbackLeibler numba gradient ignoring mask
    • show1D slice_list parameter can now be of type tuple
    • Deprecated Algorithm's max_iteration, log_file, **kwargs, max_iteration_stop_criterion, objective_to_string, verbose_output, verbose_header, run(print_interval)
    • Added optimisation.utilities.callbacks
      • Added Callback (abstract base class), ProgressCallback, TextProgressCallback, LogfileCallback
      • Deprecated Algorithm.run(callback: Callable)
      • Added Algorithm.run(callbacks: list[Callback])
    • Allow CentreOfRotationCorrector.xcorrelation to accept a list of two values in projection_index to use for the correlation
    • Bug fix in CentreOfRotationCorrector.xcorrelation finding correlation angles for limited angle data
    • New unit tests have been implemented for operators and functions to check for in place errors and the behaviour of out.
    • Bug fix for missing factor of 1/2 in SIRT update objective and catch in place errors in the SIRT constraint
    • Bug fix to allow safe in place calculation for the soft shrinkage algorithm
    • Allow Masker to take integer arrays in addition to boolean
    • Add remote data class to example data to enable download of relevant datasets from remote repositories
    • Improved import error/warning messages
    • New adjoint operator
    • Bug fix for complex matrix adjoint
    • Removed the following code which was deprecated since v22.0.0:
      • info parameter in cil.optimisation.functions.TotalVariation
      • sinogram_geometry and volume_geometry parameters in cil.plugins.astra.processors.FBP and in cil.plugins.tigre.processors.FBP
      • aquisition_geometry (misspelled) parameter in cil.plugins.tigre.processors.ProjectionOperator
      • FBP kwarg (and thus all kwargs) in cil.processors.CentreOfRotationCorrector and cil.processors.CofR_image_sharpness
      • TXRMDataReader
    • Added the ApproximateGradientSumFunction and SGFunction to allow for stochastic gradient algorithms to be created using functions with an approximate gradient and deterministic algorithms

  • 23.1.0

    • Fix bug in IndicatorBox proximal_conjugate
    • Allow CCPi Regulariser functions for non CIL object
    • Add norm for CompositionOperator
    • Refactor SIRT algorithm to make it more computationally and memory efficient
    • Optimisation in L2NormSquared
    • Added support for partitioner, when partitions have size 1
    • Fix for show_geometry bug for 2D data
    • FBP split processing bug fix - now respects panel origin set in geometry
    • Binner/Padder/Slicer bug fix - now respects panel origin set in geometry
    • Added warmstart capability to proximal evaluation of the CIL TotalVariation function.
    • Bug fix in the LinearOperator norm with an additional flag for the algorithm linearOperator.PowerMethod
    • Tidied up documentation in the framework folder

  • 23.0.1

    • Fix bug with NikonReader requiring ROI to be set in constructor.

  • 23.0.0

    • Partitioner is now able to create batches even if angle is not the outer dimension
    • Renamed max_iteration_stop_cryterion method in the Algorithm class to max_iteration_stop_criterion
    • Removed (previously deprecated) very_verbose parameter in Algorithm's run method.
    • Removed (previously deprecated) axpby method in DataContainer.
    • Deprecate use of integer compression in NEXUSDataWriter.
    • Improved and tidied up documentation for all readers and writers, including hiding special members.
    • Use arguments instead of kwargs in all readers and writers with multiple kwargs, making documentation easier.
    • Update Apache2 License Headers.

  • 22.2.0

    • BlockGeometry is iterable
    • Added partition to AcquisitionData to partition the data with 3 methods: sequential, staggered and random_permutation
    • TIGRE and ASTRA ProjectionOperator now support BlockGeometry as acquisition_geometry parameter, returning a BlockOperator
    • Added pre-set filters for recon.FBP and recon.FDK. Filters now include ram-lak, hamming, hann, cosine, shepp-logan.
    • Added RAWFileWriter to export data containers to raw files
    • Extended IndicatorBox to behave as IndicatorBoxPixelwise by passing masks in lower and upper bounds
    • Implemented IndicatorBox in numba and numpy
    • Dropped support for Python 3.6 and NumPy 1.15
    • Jenkins PR tests on Python 3.8 and NumPy 1.20
    • added yml file to create test environment
    • LeastSquares fixed docstring and unified gradient code when out is passed or not.
    • Add compression to 8bit and 16bit to TIFFWriter
    • Added convenience centre of rotation methods to AcquisitionGeometry class.
      • get_centre_of_rotation() calculates the centre of rotation of the system
      • set_centre_of_rotation() sets the system centre of rotation with an offset and angle
      • set_centre_of_rotation_by_slice() sets the system centre of rotation with offsets from two slices
    • Binner processor reworked:
      • Significant speed increase available via the C++ backend
      • Returned geometry is correctly offset where binning/cropping moves the origin
    • Slicer refactoring
      • Returned geometry is correctly offset where slicing/cropping moves the origin
    • Padder refactoring
      • Returned geometry is correctly offset where padding moves the origin
    • Github Actions:
      • update test python and numpy versions to 3.9 and 1.22
      • Update conda build action to v1.4.4
      • Fixes actions to run on ubuntu-20.04
      • Update version of upload_artifact github action to version 3.1.1
      • Update version of download_artifact github action to version 3.0.1
      • Update version of checkout github action to version 3.1.0
      • Update build-sphinx action to version 0.1.3
    • io.utilities.HDF5_utilities Added utility functions to browse hdf5 files and read datasets into numpy array
    • Implemented the analytical norm for GradientOperator
    • Added ImageData.apply_circular_mask method to mask out detector edge artefacts on reconstructed volumes
    • ROI selection, aspect ratio toggle and Play widget added to islicer
    • Add show1D display utility

  • 22.1.0

    • use assert_allclose in test_DataContainer
    • added multiple colormaps to show2D
    • Fix segfault in GradientOperator due to parameter overflows on windows systems
    • Fix angle display precision and matplotlib warning for sinograms with show2D

  • 22.0.0

    • Strongly convex functionality in TotalVariation and FGP_TV Functions
    • Refactored KullbackLeibler function class. Fix bug on gradient method for SIRF objects
    • Numba added as a CIL requirement
    • Simplify initialisation of CentreOfRotation.ImageSharpness with new parameter backend
    • Added ISTA algorithm. Improve inheritance of proximal gradient algorithms
    • Updated interface to plugins.tigre/plugins.astra FBP and ProjectionOperator classes
    • Update NikonDataReader to parse and set up geometry with: ObjectTilt CentreOfRotationTop and CentreOfRotationBottom
    • Cleaned up unit test structure and output
    • Removal of deprecated code:
      • AcquisitionGeometry __init__ no longer returns a configured geometry, use factory create methods instead
      • subset method removed, use get_slice or reorder methods
      • NikonDataReader normalize kwarg removed, use normalise
      • Algorithms initialisation x_init kwarg removed, use initial
      • Removed deprecated numpy calls
    • DataProcessors use weak-reference to input data
    • Merged CIL-ASTRA code in to CIL repository simplifying test, build and install procedures
      • Modules not moved should be considered deprecated
      • CIL remains licensed as APACHE-2.0
      • Minor bug fixes to the CPU 2D Parallel-beam FBP
    • Add ndim property for DataContainer class
    • Fixes show_geometry compatibility issue with matplotlib 3.5
    • Added ZEISSDataReader with cone/parallel beam, slicing, TXM Functionality
    • Raise exception if filename or data haven't been set in NexusDataWriter
    • Fixes error when update_objective_interval is set to 0 in an algorithm run
    • Deprecated:
      • TXRMDataReader is deprecated in favour of ZEISSDataReader
    • GitHub Actions:
      • Update to version 0.1.1 of lauramurgatroyd/build-sphinx-action for building the documentation - ensures docs are always built from cil master

  • 21.4.1

    • Removed prints from unittests and cleanup of unittest code.
    • CMake:
      • install script re-allows selection of non default install directory (#1246)
    • TIFF writer uses logging
    • Added unittests for TIFF functionality

  • 21.4.0

    • PEP 440 compliant version
    • CMake fix due to use of pip install.
    • Recon.FBP allows 'astra' backend
    • Fixed PowerMethod for square/non-square, complex/float matrices with stopping criterion.
    • CofR image_sharpness improved for large datasets
    • Geometry alignment fix for 2D datasets
    • CGLS update for sapyb to enable complex data, bugfix in use of initial
    • added sapyb and deprecated axpby. All algorithm updated to use sapyb.
    • Allow use of square brackets in file paths to TIFF and Nikon datasets

  • 21.3.1

    • Added matplotlib version dependency to conda recipe
    • Fixed TIGRE wrappers for geometry with a virtual detector
    • Fixed TIGRE wrappers for cone-beam geometry with tilted rotation axis

  • 21.3.0

    • Accelerated PDHG which handles strong convexity of functions
    • TotalVariation Function handles SIRF ImageData
    • Simulated datasets and volume added to DataExamples
    • TIGRE wrappers for parallel-beam geometry added
    • NEXUSWriter and NEXUSReader offer (8bit and 16bit) compression of data
    • show2D show_geom now return an object that can be saved with a save method
    • GradientOperator can be now used with SIRF DataContainers, both PET and MR
    • Add anisotropy in TotalVariation function
    • CCPi Regularisation plugin is refactored, only FGP_TV, FGP_dTV, TGV and TNV are exposed. Docstrings and functionality unit tests are added. Tests of the functions are meant to be in the CCPi-Regularisation toolkit itself.
    • Add dtype for ImageGeometry, AcquisitionGeometry, VectorGeometry, BlockGeometry
    • Fix GradientOperator to handle pseudo 2D CIL geometries
    • Created recon module with FBP and FDK using fast filtering library and TIGRE backprojectors
    • Added Intel IPP based library for filtering step of FBP
    • PDHG memory optimisation
    • ScaledFunction memory Optimisation
    • The github actions are merged into one action with multiple jobs
    • The conda build job uploads an artifact of the build tar.bz file which is later used by the documentation build job - which installs the package into a miniconda environment.
    • Documentation pages for recon, astra and cil-plugins are published.

  • 21.2.0

    • add version string from git describe
    • add CCPi-Regularisation toolkit in unittests
    • show_geometry implemented to display AcquisitionGeometry objects, can be imported from utilities.display
    • CentreOfRotationCorrector.image_sharpness implemented which finds the rotation axis offset by maximising sharpness of a single slice reconstruction
    • Renamed CentreOfRotationCorrector.xcorr to CentreOfRotationCorrector.xcorrelation
    • Implemented Padder processor

  • 21.1.0

    • Added TomoPhantom plugin to create 2D/3D + channel ImageData phantoms based on the TomoPhantom model library
    • Fixed bug in Zeiss reader geometry direction of rotation

  • 21.0.0

    • Show2D now takes 4D datasets and slice information as input
    • TIGRE reconstruction package wrapped for cone-beam tomography
    • Datacontainers have get_slice method which returns a dataset with a single slice of the data
    • Datacontainers have reorder method which reorders the data in memory as requested, or for use with 'astra' or 'tigre'
    • Subset method has been deprecated
    • AcquisitionData and ImageData enforce requirement for a geometry on creation
    • New processors AbsorptionTransmissionConverter and TransmissionAbsorptionConverter to convert between Absorption and Transmission
    • Implemented Binner and Slicer processors
    • Implemented MaskGenerator and Masker processors

  • 20.11.2

    • fixed windows build
    • NikonDataReader converts Nikon geometry to CIL geometry from xtekct file including detector and centre-or-rotation offsets
    • NexusdataReader supports files written with old versions of NexusDataWriter

  • 20.11

    • python module renamed to cil
    • renamed Identity->IdentityOperator, Gradient->GradientOperator, SymmetrisedGradient->SymmetrisedGradientOperator

  • 20.09.1

    • FiniteDifferenceOperator takes into consideration voxel size
    • Added CentreOfRotationCorrector
    • Removed CenterOfRotationFinder
    • moved TestData to utilities and renamed as dataexample
    • verbosity of Algorithms is independent of the update_objective_interval
    • added unittests
    • renamed
      • GradientDescent to GD
      • SparseFiniteDiff to SparseFiniteDifferenceOperator
      • LinearOperatorMatrix to MatrixOperator
    • bugfix update_objective of SPDHG

  • 20.09

    • added SPDHG algorithm
    • added TotalVariation function
    • Redesign of the AcquisitionGeometry class allowing more general acquisition trajectories than currently possible.
    • Added ZEISS reader

  • 20.04

    • Significant upgrades to the operator and function classes to allow more flexible definition of optimisation problems
    • Added multithreaded C library for calculation of finite difference and some data processing
    • Added Gradient operator using C library and numpy backends

  • 19.10

    • Improved usability with reader/writers and plotting utilities
    • Substantially improved test coverage

  • 19.07

    • Introduction of BlockFramework
    • major revision and restructuring of the whole code
    • rewritten io package

  • 19.02

    • introduction of Algorithm class
    • unit test expanded and moved to test directory
    • unified build system on Jenkins based on CCPi-VirtualMachine repo
    • switched to calendar versioning YY.0M.

  • 0.10.0

  • 0.9.4

    • Initial release