clinicadl.infer

For customizing the inference stage, such as performing post-processing or combining outputs from multiple neural networks.

Inferer

Abstract class for inferers, which define how an image is passed in a neural network during inference.

SimpleInferer

An Inferer for classical inference, i.e. when the whole image is passed in the neural network and the output is returned (with a potential postprocessing).

SlicesToImageInferer

Splits a 3D volume into 2D slices, passes them in a 2D neural network, and merges the outputs in a 3D output volume.

PatchesToImageInferer

Splits a 3D volume into 3D patches, passes them in a 3D neural network, and merges the outputs in a 3D output volume.