tsinghua-fib-lab/urbankg-knowsite

KnowSite: Leveraging Urban Knowledge Graph for Site Selection

KnowSite: Leveraging Urban Knowledge Graph for Site Selection

Requirements

To install requirements:

python 3.7.4
pytorch 1.1
dgl 0.5.1
mlflow 1.12.1

Running a model

To train (and evaluate) the model in the paper, run this command:

python main.py --dataset beijing --batch_size 128 --edim 64 --lr 0.001 --dr 0.995 --comp_opn rnn --dr 0.995 --freeze true --gcn_dropout 0.3 --kg_name kg --n_layer 2 --opn rotate --pretrain true --psa_alpha 0.5
python main.py --dataset shanghai --batch_size 128 --edim 64 --lr 0.003 --dr 0.995 --comp_opn rnn --dr 0.995 --freeze true --gcn_dropout 0.1 --kg_name kg --n_layer 2 --opn rotate --pretrain true --psa_alpha 0.8

Reference

@inproceddings{liu2023knowsite,
title 	  = {KnowSite: Leveraging Urban Knowledge Graph for Site Selection},
author	  = {Liu, Yu and Ding, Jingtao and Li, Yong},
booktitle = {ACM SIGSPATIAL},
year      = {2023}}

urbankg-knowsite's Issues

Data Visualization

Wonder how could I visualize the region and POI venue, can't read anything from the source data.