How can i input a 2-d extracted features by wavelet scattering to a deep learning network

dear all,

please i ask to a way that enable me to add a layer that hold the extracted features from wavelet scattering to a deep network?

the input in the deep network usually an input layer that hold the image datastore

[yanqi liu]

may be use feature as input 4-D matrix，and use Layers to train it

if possible，may be upload your data to analysis