What are the meanings of the metrics implemented in the /generation/decoder.py?

[WhatAShot]

What are the meanings of the metrics (cls_acc, icls_acc, topo_acc, assm_acc) implemented in the /generation/decoder.py?

[wengong-jin]

cls_acc is the motif prediction accuracy. icls_acc is the attachment prediction accuracy. assm_acc is the graph prediction accuracy. topo_acc is the accuracy of predicting when the decoder should backtrace (i.e., there will be no new neighbors for this motif node)