Thanks to visit codestin.com
Credit goes to github.com

Skip to content
/ GSLI25 Public

The src for Paper "Graph Structure Learning for Spatial-Temporal Imputation: Adapting to Node and Feature Scales"

Notifications You must be signed in to change notification settings

GSLI25/GSLI25

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Graph Structure Learning for Spatial-Temporal Imputation: Adapting to Node and Feature Scales

File Structure

  • Code: source code of our implementation
  • Data: some source files of datasets used in experiments
  • Appendix.pdf:
    • The motivation for introducing prominence modeling in graph structure learning;
    • Complete proofs for all the theoretical results in the manuscript, including Propositions 1, and Propositions 2;
    • Time complexity and space complexity analysis of our proposed node-scale spatial learning and feature-scale spatial learning modules;
    • More supplementary experiments that demonstrate the effectiveness and rationality of our method.

Preprocessing each dataset

  1. Enter the "Code" folder

  2. To get the DutchWind dataset:

python preprocessing_dutchwind.py
python preprocessing_dutchwind_getadj.py
  1. To get the BeijingMEO dataset:
python preprocessing_beijingmeo.py
python preprocessing_beijingmeo_getadj.py
  1. To get the LondonAQ dataset:
python preprocessing_londonaq.py
python preprocessing_londonaq_getadj.py
  1. To get the CN dataset:
python preprocessing_cn.py
python preprocessing_cn_getadj.py
  1. To get the Los dataset:
python preprocessing_los.py
  1. To get the LuohuTaxi dataset:
python preprocessing_luohutaxi.py

Demo Script Running

python A_main.py

Demo Forecasting Script Running

python A_main_forecasting.py

Dataset Sources

About

The src for Paper "Graph Structure Learning for Spatial-Temporal Imputation: Adapting to Node and Feature Scales"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages