TSP and CVRP

该数据集包含了为旅行商问题（TSP）和有容量限制的车辆路径问题（CVRP）生成的实例，节点数量分别为20、50和100。城市位置的坐标是从均匀分布中抽取的，对于CVRP问题，还指定了车辆容量和需求。该数据集使用二维欧几里得距离生成实例，旨在最小化总旅行距离，并包含了针对CVRP以及其他四个提及的路径问题的特定设置。数据集的规模分为20、50和100个节点，其任务旨在解决车辆路径问题。This dataset comprises instances generated for the Traveling Salesman Problem (TSP) and the Capacitated Vehicle Routing Problem (CVRP), with node counts of 20, 50, and 100 respectively. The coordinates of city locations are sampled from a uniform distribution. For CVRP instances, vehicle capacities and customer demands are also specified. All instances are generated using two-dimensional Euclidean distance, with the objective of minimizing total travel distance. The dataset includes specific configurations for CVRP and four other mentioned routing problems, and is categorized into three scales based on the number of nodes: 20, 50, and 100. Its core task is to solve vehicle routing problems.

搜集汇总

背景与挑战

背景概述

该数据集是用于解决车辆路径问题（如TSP和CVRP）的机器学习模型MDAM的代码仓库，包含多解码器注意力模型实现，提供训练、数据生成和评估脚本，依赖Python和PyTorch框架，基于学术论文构建。