search

EDAPINENUT / CLCRN

This is an official Pytorch implementation of Conditional Local Convolution for Spatio-temporal Meteorological Forecasting, AAAI 2022
130 stars 29 forks source link

visualize the spatial distribution of predictive variables #2

Closed baiyissp112 closed 2 years ago

baiyissp112 commented 2 years ago

Thank you very much for opening your source code.

How can I visualize the spatial distribution of predictive variables like temperature.

like this picture image

EDAPINENUT commented 2 years ago

First, the prediction and ground truth should be stored. After training, use `_get_time_prediction' function in Line 372 of https://github.com/BIRD-TAO/CLCRN/blob/main/supervisor.py to save them. After the prediction and the ground truth are saved, you can use visualization tools, such as geoviews. For example, you can refer to https://www.kaggle.com/code/glarvor/dataviz-explore-meteonet-using-geoviews for more details.