De-normalization of forecasted values Metr-LA dataset.

benedekrozemberczki / pytorch_geometric_temporal

PyTorch Geometric Temporal: Spatiotemporal Signal Processing with Neural Machine Learning Models (CIKM 2021)

MIT License

2.58k stars 367 forks source link

Open GreenPatterns opened 1 month ago

GreenPatterns commented 1 month ago

Hi, How can you denormalize and bring the forecasted results(batch, node=207, time=12) into the Metr_LA dataset's original range (0-70)?