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
When a dataset is updated/fixed for some reason, like was done with CDFZRAP (#1068), we tend to keep the old dataset and create a dataset with a new name for the new version, e.g. we now have CDFZRAP and CDFZRAP_NEW. This makes sense for reproducibility of old results, but I suppose there's nothing to stop someone unintentionally using the old version. I was therefore thinking that we could perhaps add some warnings to the code when such datasets are used? I'm not sure how comprehensive these warnings could/should be, but the obvious place would be in vp-setupfit. Just something like "CDFZRAP is a deprecated dataset, which exists for reproducibility. The up-to-date version is CDFZRAP_NEW". Thoughts?
The text was updated successfully, but these errors were encountered:
When a dataset is updated/fixed for some reason, like was done with CDFZRAP (#1068), we tend to keep the old dataset and create a dataset with a new name for the new version, e.g. we now have CDFZRAP and CDFZRAP_NEW. This makes sense for reproducibility of old results, but I suppose there's nothing to stop someone unintentionally using the old version. I was therefore thinking that we could perhaps add some warnings to the code when such datasets are used? I'm not sure how comprehensive these warnings could/should be, but the obvious place would be in vp-setupfit. Just something like "CDFZRAP is a deprecated dataset, which exists for reproducibility. The up-to-date version is CDFZRAP_NEW". Thoughts?
The text was updated successfully, but these errors were encountered: