[Paper] Learning the Evolutionary and Multi-scale Graph Structure for Multivariate Time Series Forecasting

openclimatefix / graph_weather

PyTorch implementation of Ryan Keisler's 2022 "Forecasting Global Weather with Graph Neural Networks" paper (https://arxiv.org/abs/2202.07575)

MIT License

179 stars 44 forks source link

[Paper] Learning the Evolutionary and Multi-scale Graph Structure for Multivariate Time Series Forecasting #69

Open jacobbieker opened 1 year ago

jacobbieker commented 1 year ago

Arxiv/Blog/Paper Link

https://arxiv.org/pdf/2206.13816.pdf

Detailed Description

Uses GNN for time series forecasting, including of solar power data. Not directly weather, but putting here as GNN based.

Context

Could be good benchmark/interesting to try out. @thomasarmstrong98 found it!

Seems like might be interesting for the new national model, or benchmarking against GSP forecasts? @JackKelly @dfulu

JackKelly commented 1 year ago

Seems like might be interesting for the new national model

Definitely sounds interesting, thanks for flagging!

To start with, I'm going to keep the new national model relatively simple (probably using a Temporal Fusion Transformer). But, you're right, if we have time then it'd be interesting to compare against this "Learning the Evolutionary and Multi-scale Graph Structure for Multivariate Time Series Forecasting" paper! I've added this paper to OCF's reference library on Zotero.