w4nderlust
commented
4 years ago In that example it suffices to keep track of mean and std in a variable to denormalize. A more generic solution would be to have, for each data preprocessing function (tokenization, normalization, etc.) a counterpart for postprocessing. There is already an open card in the project to deal with this: https://github.com/uber/ludwig/projects/1#card-33325338
In the example https://uber.github.io/ludwig/examples/#time-series-forecasting-weather-data-example the generated temperature_predictions.csv file has the values in the normalized form, how to denormalize them?
Originally posted by @InosRahul in https://github.com/uber/ludwig/issues/124#issuecomment-562135167