0.8.0

Added

• Overview of new features in v0.8
• Network modules for differentiable neural network topology search (DiNTS)
• Multiple Instance Learning transforms and models for digital pathology WSI analysis
• Vision transformers for self-supervised representation learning
• Contrastive loss for self-supervised learning
• Finalized major improvements of 200+ components in monai.transforms to support input and backend in PyTorch and NumPy
• Initial registration module benchmarking with GlobalMutualInformationLoss as an example
• monai.transforms documentation with visual examples and the utility functions
• Event handler for MLfLow integration
• Enhanced data visualization functions including blend_images and matshow3d
• RandGridDistortion and SmoothField in monai.transforms
• Support of randomized shuffle buffer in iterable datasets
• Performance review and enhancements for data type casting
• Cumulative averaging API with distributed environment support
• Module utility functions including require_pkg and pytorch_after
• Various usability enhancements such as allow_smaller when sampling ROI and wrap_sequence when casting object types
• tifffile support in WSIReader
• Regression tests for the fast training workflows
• Various tutorials and demos including educational contents at MONAI Bootcamp 2021

Changed

• Base Docker image upgraded to nvcr.io/nvidia/pytorch:21.10-py3 from nvcr.io/nvidia/pytorch:21.08-py3
• Decoupled TraceKeys and TraceableTransform APIs from InvertibleTransform
• Skipping affine-based resampling when resample=False in NiftiSaver
• Deprecated threshold_values: bool and num_classes: int in AsDiscrete
• Enhanced apply_filter for spatially 1D, 2D and 3D inputs with non-separable kernels
• Logging with logging in downloading and model archives in monai.apps
• API documentation site now defaults to stable instead of latest
• skip-magic-trailing-comma in coding style enforcements
• Pre-merge CI pipelines now include unit tests with Nvidia Ampere architecture

Removed

• Support for PyTorch 1.5
• The deprecated DynUnetV1 and the related network blocks
• GitHub self-hosted CI/CD pipelines for package releases

Fixed

• Support of path-like objects as file path inputs in most modules
• Issue of decollate_batch for dictionary of empty lists
• Typos in documentation and code examples in various modules
• Issue of no available keys when allow_missing_keys=True for the MapTransform
• Issue of redundant computation when normalization factors are 0.0 and 1.0 in ScaleIntensity
• Incorrect reports of registered readers in ImageReader
• Wrong numbering of iterations in StatsHandler
• Naming conflicts in network modules and aliases
• Incorrect output shape when reduction="none" in FocalLoss
• Various usability issues reported by users

0.7.0

Added

• Overview of new features in v0.7
• Initial phase of major usability improvements in monai.transforms to support input and backend in PyTorch and NumPy
• Performance enhancements, with profiling and tuning guides for typical use cases
• Reproducing training modules and workflows of state-of-the-art Kaggle competition solutions
• 24 new transforms, including
  • OneOf meta transform
  • DeepEdit guidance signal transforms for interactive segmentation
  • Transforms for self-supervised pre-training
  • Integration of NVIDIA Tools Extension (NVTX)
  • Integration of cuCIM
  • Stain normalization and contextual grid for digital pathology
• Transchex network for vision-language transformers for chest X-ray analysis
• DatasetSummary utility in monai.data
• WarmupCosineSchedule
• Deprecation warnings and documentation support for better backwards compatibility
• Padding with additional kwargs and different backend API
• Additional options such as dropout and norm in various networks and their submodules

Changed

• Base Docker image upgraded to nvcr.io/nvidia/pytorch:21.08-py3 from nvcr.io/nvidia/pytorch:21.06-py3
• Deprecated input argument n_classes, in favor of num_classes
• Deprecated input argument dimensions and ndims, in favor of spatial_dims
• Updated the Sphinx-based documentation theme for better readability
• NdarrayTensor type is replaced by NdarrayOrTensor for simpler annotations
• Self-attention-based network blocks now support both 2D and 3D inputs

Removed

• The deprecated TransformInverter, in favor of monai.transforms.InvertD
• GitHub self-hosted CI/CD pipelines for nightly and post-merge tests
• monai.handlers.utils.evenly_divisible_all_gather
• monai.handlers.utils.string_list_all_gather

Fixed

• A Multi-thread cache writing issue in LMDBDataset
• Output shape convention inconsistencies of the image readers
• Output directory and file name flexibility issue for NiftiSaver, PNGSaver
• Requirement of the label field in test-time augmentation
• Input argument flexibility issues for ThreadDataLoader
• Decoupled Dice and CrossEntropy intermediate results in DiceCELoss
• Improved documentation, code examples, and warning messages in various modules
• Various usability issues reported by users

0.6.0

Added

• Overview document for feature highlights in v0.6
• 10 new transforms, a masked loss wrapper, and a NetAdapter for transfer learning
• APIs to load networks and pre-trained weights from Clara Train Medical Model ARchives (MMARs)
• Base metric and cumulative metric APIs, 4 new regression metrics
• Initial CSV dataset support
• Decollating mini-batch as the default first postprocessing step
• Initial backward compatibility support via monai.utils.deprecated
• Attention-based vision modules and UNETR for segmentation
• Generic module loaders and Gaussian mixture models using the PyTorch JIT compilation
• Inverse of image patch sampling transforms
• Network block utilities get_[norm, act, dropout, pool]_layer
• unpack_items mode for apply_transform and Compose
• New event INNER_ITERATION_STARTED in the deepgrow interactive workflow
• set_data API for cache-based datasets to dynamically update the dataset content
• Fully compatible with PyTorch 1.9
• --disttests and --min options for runtests.sh
• Initial support of pre-merge tests with Nvidia Blossom system

Changed

• Base Docker image upgraded to nvcr.io/nvidia/pytorch:21.06-py3 from
  nvcr.io/nvidia/pytorch:21.04-py3
• Optionally depend on PyTorch-Ignite v0.4.5 instead of v0.4.4
• Unified the demo, tutorial, testing data to the project shared drive, and
  Project-MONAI/MONAI-extra-test-data
• Unified the terms: post_transform is renamed to postprocessing, pre_transform is renamed to preprocessing
• Unified the postprocessing transforms and event handlers to accept the "channel-first" data format
• evenly_divisible_all_gather and string_list_all_gather moved to monai.utils.dist

Removed

• Support of 'batched' input for postprocessing transforms and event handlers
• TorchVisionFullyConvModel
• set_visible_devices utility function
• SegmentationSaver and TransformsInverter handlers

Fixed

• Issue of handling big-endian image headers
• Multi-thread issue for non-random transforms in the cache-based datasets
• Persistent dataset issue when multiple processes sharing a non-exist cache location
• Typing issue with Numpy 1.21.0
• Loading checkpoint with both model and optmizier using CheckpointLoader when strict_shape=False
• SplitChannel has different behaviour depending on numpy/torch inputs
• Transform pickling issue caused by the Lambda functions
• Issue of filtering by name in generate_param_groups
• Inconsistencies in the return value types of class_activation_maps
• Various docstring typos
• Various usability enhancements in monai.transforms

0.5.3

Changed

• Project default branch renamed to dev from master
• Base Docker image upgraded to nvcr.io/nvidia/pytorch:21.02-py3 from nvcr.io/nvidia/pytorch:21.04-py3
• Enhanced type checks for the iteration_metric handler
• Enhanced PersistentDataset to use tempfile during caching computation
• Enhanced various info/error messages
• Enhanced performance of RandAffine
• Enhanced performance of SmartCacheDataset
• Optionally requires cucim when the platform is Linux
• Default device of TestTimeAugmentation changed to cpu

Fixed

• Download utilities now provide better default parameters
• Duplicated key_transforms in the patch-based transforms
• A multi-GPU issue in ClassificationSaver
• A default meta_data issue in SpacingD
• Dataset caching issue with the persistent data loader workers
• A memory issue in permutohedral_cuda
• Dictionary key issue in CopyItemsd
• box_start and box_end parameters for deepgrow SpatialCropForegroundd
• Tissue mask array transpose issue in MaskedInferenceWSIDataset
• Various type hint errors
• Various docstring typos

Added

• Support of to_tensor and device arguments for TransformInverter
• Slicing options with SpatialCrop
• Class name alias for the networks for backward compatibility
• k_divisible option for CropForeground
• map_items option for Compose
• Warnings of inf and nan for surface distance computation
• A print_log flag to the image savers
• Basic testing pipelines for Python 3.9

0.5.0

Added

• Overview document for feature highlights in v0.5.0
• Invertible spatial transforms
  • InvertibleTransform base APIs
  • Batch inverse and decollating APIs
  • Inverse of Compose
  • Batch inverse event handling
  • Test-time augmentation as an application
• Initial support of learning-based image registration:
  • Bending energy, LNCC, and global mutual information loss
  • Fully convolutional architectures
  • Dense displacement field, dense velocity field computation
  • Warping with high-order interpolation with C++/CUDA implementations
• Deepgrow modules for interactive segmentation:
  • Workflows with simulations of clicks
  • Distance-based transforms for guidance signals
• Digital pathology support:
  • Efficient whole slide imaging IO and sampling with Nvidia cuCIM and SmartCache
  • FROC measurements for lesion
  • Probabilistic post-processing for lesion detection
  • TorchVision classification model adaptor for fully convolutional analysis
• 12 new transforms, grid patch dataset, ThreadDataLoader, EfficientNets B0-B7
• 4 iteration events for the engine for finer control of workflows
• New C++/CUDA extensions:
  • Conditional random field
  • Fast bilateral filtering using the permutohedral lattice
• Metrics summary reporting and saving APIs
• DiceCELoss, DiceFocalLoss, a multi-scale wrapper for segmentation loss computation
• Data loading utilities：
  • decollate_batch
  • PadListDataCollate with inverse support
• Support of slicing syntax for Dataset
• Initial Torchscript support for the loss modules
• Learning rate finder
• Allow for missing keys in the dictionary-based transforms
• Support of checkpoint loading for transfer learning
• Various summary and plotting utilities for Jupyter notebooks
• Contributor Covenant Code of Conduct
• Major CI/CD enhancements covering the tutorial repository
• Fully compatible with PyTorch 1.8
• Initial nightly CI/CD pipelines using Nvidia Blossom Infrastructure

Changed

• Enhanced list_data_collate error handling
• Unified iteration metric APIs
• densenet* extensions are renamed to DenseNet*
• se_res* network extensions are renamed to SERes*
• Transform base APIs are rearranged into compose, inverse, and transform
• _do_transform flag for the random augmentations is unified via RandomizableTransform
• Decoupled post-processing steps, e.g. softmax, to_onehot_y, from the metrics computations
• Moved the distributed samplers to monai.data.samplers from monai.data.utils
• Engine's data loaders now accept generic iterables as input
• Workflows now accept additional custom events and state properties
• Various type hints according to Numpy 1.20
• Refactored testing utility runtests.sh to have --unittest and --net integration tests options
• Base Docker image upgraded to nvcr.io/nvidia/pytorch:21.02-py3 from nvcr.io/nvidia/pytorch:20.10-py3
• Docker images are now built with self-hosted environments
• Primary contact email updated to [email protected]
• Now using GitHub Discussions as the primary communication forum

Removed

• Compatibility tests for PyTorch 1.5.x
• Format specific loaders, e.g. LoadNifti, NiftiDataset
• Assert statements from non-test files
• from module import * statements, addressed flake8 F403

Fixed

• Uses American English spelling for code, as per PyTorch
• Code coverage now takes multiprocessing runs into account
• SmartCache with initial shuffling
• ConvertToMultiChannelBasedOnBratsClasses now supports channel-first inputs
• Checkpoint handler to save with non-root permissions
• Fixed an issue for exiting the distributed unit tests
• Unified DynUNet to have single tensor output w/o deep supervision
• SegmentationSaver now supports user-specified data types and a squeeze_end_dims flag
• Fixed *Saver event handlers output filenames with a data_root_dir option
• Load image functions now ensure little-endian
• Fixed the test runner to support regex-based test case matching
• Usability issues in the event handlers

0.4.0

Added

• Overview document for feature highlights in v0.4.0
• Torchscript support for the net modules
• New networks and layers:
  • Discrete Gaussian kernels
  • Hilbert transform and envelope detection
  • Swish and Mish activation
  • Acti-norm-dropout block
  • Upsampling layer
  • Autoencoder, Variational autoencoder
  • FCNet
• Support of initialisation from pre-trained weights for densenet, SENet, multichannel AHNet
• Layer-wise learning rate API
• New model metrics and event handlers based on occlusion sensitivity, confusion matrix, surface distance
• CAM/GradCAM/GradCAM++
• File format-agnostic image loader APIs with Nibabel, ITK readers
• Enhancements for dataset partition, cross-validation APIs
• New data APIs:
  • LMDB-based caching dataset
  • Cache-N-transforms dataset
  • Iterable dataset
  • Patch dataset
• Weekly PyPI release
• Fully compatible with PyTorch 1.7
• CI/CD enhancements:
  • Skip, speed up, fail fast, timed, quick tests
  • Distributed training tests
  • Performance profiling utilities
• New tutorials and demos:
  • Autoencoder, VAE tutorial
  • Cross-validation demo
  • Model interpretability tutorial
  • COVID-19 Lung CT segmentation challenge open-source baseline
  • Threadbuffer demo
  • Dataset partitioning tutorial
  • Layer-wise learning rate demo
  • MONAI Bootcamp 2020

Changed

• Base Docker image upgraded to nvcr.io/nvidia/pytorch:20.10-py3 from nvcr.io/nvidia/pytorch:20.08-py3

Backwards Incompatible Changes

• monai.apps.CVDecathlonDataset is extended to a generic monai.apps.CrossValidation with an dataset_cls option
• Cache dataset now requires a monai.transforms.Compose instance as the transform argument
• Model checkpoint file name extensions changed from .pth to .pt
• Readers' get_spatial_shape returns a numpy array instead of list
• Decoupled postprocessing steps such as sigmoid, to_onehot_y, mutually_exclusive, logit_thresh from metrics and event handlers, the postprocessing steps should be used accordingly before computing the metrics
• ConfusionMatrixMetric and DiceMetric computation now returns an additional not_nans flag to indicate valid results
• UpSample optional mode supports "deconv", "nontrainable", "pixelshuffle"; interp_mode is only used when mode is "nontrainable"
• SegResNet optional upsample_mode now supports "deconv", "nontrainable", "pixelshuffle"
• monai.transforms.Compose class inherits monai.transforms.Transform
• In Rotate, Rotated, RandRotate, RandRotated transforms, the angle related parameters are interpreted as angles in radians instead of degrees.
• SplitChannel and SplitChanneld moved from transforms.post to transforms.utility

Removed

• Support of PyTorch 1.4

Fixed

• Enhanced the Dice related loss functions for stability and flexibility
• Sliding window inference memory and device issues
• Revised transforms:
  • Normalize intensity datatype and normalizer types
  • Padding modes for zoom
  • Crop returns coordinates
  • Select items transform
  • Weighted patch sampling
  • Option to keep aspect ratio for zoom
• Various CI/CD issues

0.3.0

Added

• Overview document for feature highlights in v0.3.0
• Automatic mixed precision support
• Multi-node, multi-GPU data parallel model training support
• 3 new evaluation metric functions
• 11 new network layers and blocks
• 6 new network architectures
• 14 new transforms, including an I/O adaptor
• Cross validation module for DecathlonDataset
• Smart Cache module in dataset
• monai.optimizers module
• monai.csrc module
• Experimental feature of ImageReader using ITK, Nibabel, Numpy, Pillow (PIL Fork)
• Experimental feature of differentiable image resampling in C++/CUDA
• Ensemble evaluator module
• GAN trainer module
• Initial cross-platform CI environment for C++/CUDA code
• Code style enforcement now includes isort and clang-format
• Progress bar with tqdm

Changed

• Now fully compatible with PyTorch 1.6
• Base Docker image upgraded to nvcr.io/nvidia/pytorch:20.08-py3 from nvcr.io/nvidia/pytorch:20.03-py3
• Code contributions now require signing off on the Developer Certificate of Origin (DCO)
• Major work in type hinting finished
• Remote datasets migrated to Open Data on AWS
• Optionally depend on PyTorch-Ignite v0.4.2 instead of v0.3.0
• Optionally depend on torchvision, ITK
• Enhanced CI tests with 8 new testing environments

Removed

• MONAI/examples folder (relocated into Project-MONAI/tutorials)
• MONAI/research folder (relocated to Project-MONAI/research-contributions)

Fixed

• dense_patch_slices incorrect indexing
• Data type issue in GeneralizedWassersteinDiceLoss
• ZipDataset return value inconsistencies
• sliding_window_inference indexing and device issues
• importing monai modules may cause namespace pollution
• Random data splits issue in DecathlonDataset
• Issue of randomising a Compose transform
• Various issues in function type hints
• Typos in docstring and documentation
• PersistentDataset issue with existing file folder
• Filename issue in the output writers

0.2.0

Added

• Overview document for feature highlights in v0.2.0
• Type hints and static type analysis support
• MONAI/research folder
• monai.engine.workflow APIs for supervised training
• monai.inferers APIs for validation and inference
• 7 new tutorials and examples
• 3 new loss functions
• 4 new event handlers
• 8 new layers, blocks, and networks
• 12 new transforms, including post-processing transforms
• monai.apps.datasets APIs, including MedNISTDataset and DecathlonDataset
• Persistent caching, ZipDataset, and ArrayDataset in monai.data
• Cross-platform CI tests supporting multiple Python versions
• Optional import mechanism
• Experimental features for third-party transforms integration

Changed

For more details please visit the project wiki

• Core modules now require numpy >= 1.17
• Categorized monai.transforms modules into crop and pad, intensity, IO, post-processing, spatial, and utility
• Most transforms are now implemented with PyTorch native APIs
• Code style enforcement and automated formatting workflows now use autopep8 and black
• Base Docker image upgraded to nvcr.io/nvidia/pytorch:20.03-py3 from nvcr.io/nvidia/pytorch:19.10-py3
• Enhanced local testing tools
• Documentation website domain changed to https://docs.monai.io

Removed

• Support of Python < 3.6
• Automatic installation of optional dependencies including pytorch-ignite, nibabel, tensorboard, pillow, scipy, scikit-image

Fixed

• Various issues in type and argument names consistency
• Various issues in docstring and documentation site
• Various issues in unit and integration tests
• Various issues in examples and notebooks

0.1.0

Added

• Public alpha source code under the Apache 2.0 license (highlights)
• Various tutorials and examples
  • Medical image classification and segmentation workflows
  • Spacing/orientation-aware preprocessing with CPU/GPU and caching
  • Flexible workflows with external engines including PyTorch Ignite and Lightning
• Various GitHub Actions
  • CI/CD pipelines via self-hosted runners
  • Documentation publishing via readthedocs.org
  • PyPI package publishing
• Contributing guidelines
• A project logo and badges