0.4.0
版本发布时间: 2020-12-15 23:21:13
Project-MONAI/MONAI最新发布版本:1.4.0(2024-10-17 08:54:28)
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
fromnvcr.io/nvidia/pytorch:20.08-py3
Backwards Incompatible Changes
-
monai.apps.CVDecathlonDataset
is extended to a genericmonai.apps.CrossValidation
with andataset_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
andDiceMetric
computation now returns an additionalnot_nans
flag to indicate valid results -
UpSample
optionalmode
supports"deconv"
,"nontrainable"
,"pixelshuffle"
;interp_mode
is only used whenmode
is"nontrainable"
-
SegResNet
optionalupsample_mode
now supports"deconv"
,"nontrainable"
,"pixelshuffle"
-
monai.transforms.Compose
class inheritsmonai.transforms.Transform
- In
Rotate
,Rotated
,RandRotate
,RandRotated
transforms, theangle
related parameters are interpreted as angles in radians instead of degrees. -
SplitChannel
andSplitChanneld
moved fromtransforms.post
totransforms.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