Dataset Modules

his module is contains functionality to bundle and format input resources in corresponding to the specified input format into tf.data.Dataset. This is necessary to manage the streaming of large input arrays efficiently.

Summary

Currently available dataset modules:

cmrsim.datasets.BaseDataset(filter_inputs, ...)

Basic implementation of a module that constructs a tf.data.Dataset from the dictionary of numpy arrays, describing the digital phantom.

cmrsim.datasets.AnalyticDataset(array_dictionary)

cmrsim.datasets.BlochDataset(array_dictionary)

cmrsim.datasets.RefillingFlowDataset(mesh, ...)

Wraps a vtk-mesh of containing a velocity field.

cmrsim.datasets.MeshDataset(mesh, timing_ms)

This class bundles basic functionality to handles dynamic meshes and implements convenience functionality like rendering or slice selection.

cmrsim.datasets.CardiacMeshDataset(mesh, ...)

Cardiac mesh model assuming the continuous node-ordering: