we should have an open discussion about what quantities of the forecast should appear as meta data in an evaluation result file. this comes from the fact that im working with !~1750 forecasts all having different start times, and it made me wonder how much of the forecast metadata should be associated with the result .json file.
we should have an open discussion about what quantities of the forecast should appear as meta data in an evaluation result file. this comes from the fact that im working with !~1750 forecasts all having different start times, and it made me wonder how much of the forecast metadata should be associated with the result .json file.