Wish to build encoder-decoder network which accepts graphs as input as well as outputs.
I have training set of such input and output graphs, supervised manner.
Similar to Semantic Segmentation dataset, where there is an input image and corresponding output segmented image.
How to build such encoder decode network? Please note that as both, input and output are different, this can not be AutoEncoder.
More details on the original problem at: GitHub - yogeshhk/MidcurveNN: Computation of Midcurve of Thin Polygons using Neural Networks