Skip to content

Latest commit

 

History

History
255 lines (217 loc) · 16 KB

CHANGELOG.md

File metadata and controls

255 lines (217 loc) · 16 KB

Changelog

0.20.0 (2024-09-20)

  • Fix silenced exception in Queue when using faulty transform by @Zhack47 in #1101
  • Fix RescaleIntensity by @fepegar in #1116
  • Queue length modification with the use of DDP by @haughty-yeon in #1127
  • Fix get_subjects_from_batch ignoring metadata by @KonoMaxi in #1131
  • Fix random errors in queue test by @fepegar in #1142
  • Add masking_method arg name in Mask by @lchauvin in #1137
  • Enable plotting of PNG images with alpha channel by @ChristianHinge in #1150
  • Fix transforms taking strings instead of sequences by @fepegar in #1153
  • Improve docs for "isotropic" arg in RandomAffine by @AminAlam in #1163
  • Add support to slice images and subjects by @fepegar in #1170
  • Make transform return the same type as input by @haarisr in #1182
  • Fix error which copying a subclass of Subject with keyword attributes by @c-winder in #1186
  • Add SubjectsLoader by @fepegar in #1194

0.19.0 (2023-07-26)

  • Replace p by **kwargs in all transforms by @dmus in #370
  • Implementation of include/exclude args by @dmus in #373
  • Remove internal self.fill, and keep padding_mode as string or number by @romainVala in #378
  • Remove checks and casts to float32 by @dmus in #380
  • Fix conversion to Pillow image by @fepegar in #385
  • Print image path if probability map is empty by @fepegar in #386
  • Remove torchvision requirement by @fepegar in #391
  • Fix label sampler and improve error message by @fepegar in #394
  • Fix error with invertible custom transforms, also keep include/exclude lists for inverse transforms. by @efirdc in #398
  • New label transformations by @efirdc in #402
  • Add transform to ensure shape multiple of N by @fepegar in #401
  • Allow empty list as input to Compose. by @efirdc in #404
  • Cast to float before computing mean by @fepegar in #408
  • Add option to keep copies of original images by @fepegar in #413
  • Delete image attribute when removing image by @fepegar in #415
  • Add option to ignore intensity transforms when inverting by @fepegar in #417
  • Add OneHot transform by @fepegar in #420
  • Add option to modify interpolation when inverting by @fepegar in #418
  • Add some label transforms for binary morphological operations by @fepegar in #424
  • Small improvements for Queue by @fepegar in #423
  • Fix CropOrPad for 2D images by @fepegar in #435
  • Add EPISURG dataset by @fepegar in #433
  • Add list of masks functionality to HistogramStandardization.train by @Linardos in #446
  • Add support for custom image readers by @fepegar in #454
  • Fix probabilities in label sampler by @fepegar in #459
  • Use voxels rather than bounds when plotting images by @fepegar in #487
  • Add input range kwarg for RescaleIntensity by @fepegar in #499
  • Ignore NumPy underflows by @fepegar in #503
  • Remove unused imports by @deepsource-autofix in #506
  • Remove methods with unnecessary super delegation. by @deepsource-autofix in #507
  • Add examples gallery to documentation by @fepegar in #508
  • Fix 0 prob random sample in WeightedSampler by @TylerSpears in #511
  • Avoid deprecation warnings by @fepegar in #517
  • Add functions to process batches by @fepegar in #516
  • Make the grid sampler compatible with the queue by @fepegar in #520
  • Compress and cleanup downloaded datasets by @fepegar in #531
  • Set minimum version for NumPy requirement by @fepegar in #532
  • Fix unexpected exception handling by @siahuat0727 in #542
  • Fix 'unraveling -1' in the weighted random sampler. by @dvolgyes in #551
  • Add Mask transform by @Svdvoort in #527
  • Add CopyAffine transform by @albansteff in #584
  • Added Image method to convert from sitk to torchio by @mattwarkentin in #616
  • Added intensity clamp transform by @mattwarkentin in #622
  • Fix default pad value for label maps by @fepegar in #627
  • Add tolerance for numeric checks by @justusschock in #592
  • Fix syntax error for Python < 3.8 by @fepegar in #639
  • Add support to pass shape and affine to Resample by @fepegar in #640
  • Fix single slices metadata being lost by @fepegar in #641
  • Add Resize transform by @fepegar in #642
  • Raise from None to reduce error verbosity by @fepegar in #649
  • Add RSNA-MICCAI brain tumor dataset by @fepegar in #650
  • Fix visualization for subject with many images by @fepegar in #653
  • Fix error when normalization mask is empty by @fepegar in #656
  • Enable Indexing by int-Compatible type by @justusschock in #670
  • Add support to crop/pad to the mask's bounding box by @fepegar in #678
  • Add support to save GIFs by @fepegar in #680
  • Add some Visible Human datasets by @fepegar in #679
  • Add indices parameter for plotting volumes by @laynr in #683
  • Remove channels_last kwarg from Image by @fepegar in #685
  • Add parse_input kwarg to Transform by @fepegar in #696
  • Cast translation argument to float64 for SimpleITK by @fepegar in #702
  • Make Affine take RAS parameters of floating->ref by @fepegar in #712
  • Rename master to main by @fepegar in #717
  • Add method to show image with external viewer by @fepegar in #727
  • Add kwarg to not check spatial consistency by @fepegar in #735
  • Retrieve transforms history from batch dictionary by @fepegar in #745
  • Add logic to invert OneHot transform by @fepegar in #748
  • Fix wrong class when creating subjects from batch by @fepegar in #749
  • Minor quality of life functions for inspecting scalar images and label maps by @mattwarkentin in #728
  • Fix CropOrPad incorrectly storing bounds by @fepegar in #760
  • Fix affine getter for multipath images by @fepegar in #763
  • Check data type compatibility for custom reader by @fepegar in #770
  • Use PyTorch to shuffle the queue patches by @fepegar in #776
  • Fix Blur using only the first standard deviation by @fepegar in #772
  • Add support to specify the interpolation type for label images by @snavalm in #791
  • Fix overlapping in patches aggregator by @fepegar in #832
  • Add support for torch >= 1.11 by @snipdome in #838
  • Fix new SimpleITK release not being installed by @fepegar in #852
  • Fix the wrong mapping in sequential labels transform by @iamSmallY in #841
  • Return figure in plot_volume function by @dmus in #872
  • Use PyTorch to compute Fourier transforms by @fepegar in #389
  • Remove support for Python 3.6 by @fepegar in #881
  • Add support to use different numbers of samples in the queue by @fepegar in #795
  • Fix computation of kernels in Blur transform by @fepegar in #861
  • Add support to pass label keys for dict input by @fepegar in #879
  • Add support to mask 4D images with 3D masks by @fepegar in #908
  • Add fftshift to fourier_transform with torch.fft by @iimog in #912
  • Fix copying of subclasses of Subject by @justusschock in #794
  • Use tqdm.auto by @fepegar in #926
  • Add hann window function in GridAggregator as padding_mode by @LucaLumetti in #900
  • Add RSNA 2022 Cervical Spine Fracture Detection dataset by @fepegar in #943
  • Replace Click with Typer by @fepegar in #978
  • Fix number of samples per subject in queue by @fepegar in #981
  • Stop loading data when copying unloaded image by @fepegar in #982
  • Add support to pass a callable as padding mode by @mueller-franzes in #959
  • Use indexing instead of masked_select for faster normalization by @ramonemiliani93 in #1018
  • Remove Visible Human Project datasets by @fepegar in #1026
  • Add support for Python 3.11 by @fepegar in #1008
  • Make subject parameter mandatory by @fepegar in #1029
  • Improve support for queue in distributed training by @hsyang1222 in #1021
  • Stop ignoring intensity transforms when computing inverse by @fepegar in #1039
  • Add support to invert RescaleIntensity transform by @nicoloesch in #998
  • Fix HistogramStandardization example by @vedal in #1022
  • Add workflow to validate pull request title by @fepegar in #1042
  • Add support to unload images by @fepegar in #983
  • Split CI workflows by @fepegar in #1047
  • Warn if 'mean' padding mode is used for label maps by @fepegar in #1065
  • Fix unloaded image affine not read after CopyAffine by @fepegar in #1072
  • Fix custom reader not propagated when copying by @fepegar in #1091
  • Remove support for Python 3.7 by @fepegar in #1099

0.18.0 (2020-11-29)

  • Add FPG dataset
  • Optimize cropping in samplers
  • Optimize implementation of UniformSampler and WeightedSampler
  • Fix non integer labels in Colin 27 version 2008 (#223)
  • Add RandomAnisotropy transform
  • Add RandomGamma transform
  • Add ScalarImage and LabelMap classes
  • Add RandomLabelsToImage transform
  • Add support for more input types in Transform
  • Add support for 4D images
  • Add ICBM dataset
  • Add support to specify axes as anatomical labels
  • Add SpatialTransform and IntensityTransform
  • Replace ImagesDataset with SubjectsDataset
  • Add support to pass multiple paths to Image
  • Refactor reproducibility features
  • Add deterministic versions of all random transforms
  • Add support to invert transforms
  • Add support for SimpleITK > 1
  • Add pre-commit to help contributions and improve code quality
  • Add DICOM data for testing
  • Add some visualization support (Subject.plot, Image.plot)
  • Add support to pass parameters per axis, e.g. for RandomAffine (#346)
  • Remove deprecated transforms Resample and CenterCropOrPad

0.17.0 (2020-06-23)

  • Add transforms history to Subject attributes to improve traceability
  • Add support to use an initial transformation in Resample
  • Add support to use an image file as target in Resample
  • Add mean argument to RandomNoise
  • Add tensor support for transforms
  • Add support to use strings as interpolation argument
  • Add support for 2D images
  • Add attribute access to Subject and Image
  • Add MNI and 3D Slicer datasets
  • Add intensity argument to RandomGhosting
  • Add translation argument to RandomAffine
  • Add shape, spacing and orientation attributes to Image and Subject
  • Refactor samplers
  • Refactor inference classes
  • Add 3D Slicer extension
  • Add ITK-SNAP datasets
  • Add support to take NumPy arrays as transforms input
  • Optimize cropping using PyTorch
  • Optimizing transforms by reducing number of tensor copying
  • Improve representation (repr()) of Image
  • Use lazy loading in Image

0.16.0 (2020-04-21)

  • Add advanced padding options for RandomAffine
  • Add reference space options in Resample
  • Add probability argument to all transforms
  • Add OneOf and Compose transforms to improve composability

0.15.0 (2020-04-07)

  • Refactor RandomElasticDeformation transform
  • Make Subject inherit from dict

0.14.0 (2020-03-31)

  • Add datasets module
  • Add support for DICOM files
  • Add documentation
  • Add CropOrPad transform

0.13.0 (2020-02-24)

  • Add Subject class
  • Add random blur transform
  • Add lambda transform
  • Add random patches swapping transform
  • Add MRI k-space ghosting artefact augmentation

0.12.0 (2020-01-21)

  • Add ToCanonical transform
  • Add CenterCropOrPad transform

0.11.0 (2020-01-15)

  • Add Resample transform

0.10.0 (2020-01-15)

  • Add Pad transform
  • Add Crop transform

0.9.0 (2020-01-14)

  • Add CLI tool to transform an image from file

0.8.0 (2020-01-11)

  • Add Image class

0.7.0 (2020-01-02)

  • Make transforms use PyTorch tensors consistently

0.6.0 (2020-01-02)

  • Add support for NRRD

0.5.0 (2020-01-01)

  • Add bias field transform

0.4.0 (2019-12-29)

  • Add MRI k-space motion artefact augmentation

0.3.0 (2019-12-21)

  • Add Rescale transform
  • Add support for multimodal data and missing modalities

0.2.0 (2019-12-06)

  • First release on PyPI.