Code implementation of "Learning Efficient Online 3D Bin Packing on Packing Configuration Trees". We propose to enhance the practical applicability of online 3D Bin Packing Problem (BPP) via learning on a hierarchical packing configuration tree which makes the deep reinforcement learning (DRL) model easy to deal with practical constraints and well-performing even with continuous solution space.
In the original paper of GAT, features are aggregated only from neighboring nodes, but it seems in the current implementation that the aggregation is determined using all valid nodes. Is that correct? I wonder if this dilutes local information and affects the model's performance.
In the original paper of GAT, features are aggregated only from neighboring nodes, but it seems in the current implementation that the aggregation is determined using all valid nodes. Is that correct? I wonder if this dilutes local information and affects the model's performance.