You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We provide benchmark results of spatiotemporal prediction learning (STL) methods on famous weather prediction datasets, WeatherBench. More STL methods will be supported in the future. Issues and PRs are welcome! Visualization of GIF is released.
We provide visualization figures of various weather prediction methods on Weather Bench (single variable). You can plot your own visualization with tested results (e.g., work_dirs/exp_name/saved) by vis_video.py. Note that --vis_dirs denotes visualize all experimental folders under the path, and --vis_channel can select the channel for visualization. For example, run plotting with the script:
We provide temperature prediction benchmark results on the popular WeatherBench dataset (temperature prediction t2m) using $12\rightarrow 12$ frames prediction setting.
Visualization of STL Benchmarks on Temperature (t2m)
Similar to Moving MNIST Benchmarks, we benchmark STL methods training 50 epochs with single GPU on r (%). We provide config files in configs/weather/r_5_625 for 5.625 settings ($32\times 64$ resolutions).
Visualization of STL Benchmarks on Wind Component (uv10)
Similar to Moving MNIST Benchmarks, we benchmark popular STL methods training 50 epochs with single GPU on uv10 (ms-1). We provide config files in configs/weather/uv10_5_625 for 5.625 settings ($32\times 64$ resolutions). Notice that the input data of uv10 has two channels.
Visualization of STL Benchmarks on Cloud Cover (tcc)
Similar to Moving MNIST Benchmarks, we benchmark popular STL methods training 50 epochs with single GPU on tcc (%). We provide config files in configs/weather/tcc_5_625 for 5.625 settings ($32\times 64$ resolutions).