What’s new in 0.4.0¶
New operator to automate inference with the newly introduced MONAI Bundle
MONAI Bundle Inference Operator¶
The new operator,
MONAI Bundle Inference Operator, is intended to automate the inference with a MONAI Bundle in TorchScript with the following functionalities:
Parse the model metadata and extra configuration data in the TorchScript file
Instanciate MONAI transforms and inferer objects per bundle configuration data
Convert input/output of the operator to and from model network input
Support named model and can be used in a multi-model application
The example spleen segmentation application was accordingly updated to demonstrate the use of this new operator with the newly published Spleen CT Segmentation Bundle.
More details are in the tutorial.
[FEA] A generic operator to automate inference execution with a MONAI Bundle
[FEA] Multi-model support in an application built with MONAI Deploy App SDK
[DOC] Missing python dependencies for the segmentation app and the duplicate
Please visit GETTING STARTED guide and follow the tutorials.
You can learn more about SDK usage through DEVELOPING WITH SDK.
Please let us know how you like it and what could be improved by submitting an issue or asking questions