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
{{ message }}
This repository has been archived by the owner on Jun 2, 2023. It is now read-only.
Also, I think the val and tst data need to be logged here. Why are only the training data logged? The mean and standard deviation are computed on the logged data and then applied to the val and tst data, which are not logged. Some comments in the preproc_utils need to be updated to describe what log_y_vars and log_vars should be.
And we should subtract off the small constant after transforming from log to real space here.
I can address these edits in a PR
The text was updated successfully, but these errors were encountered:
jds485
changed the title
Change xr.ufuncs.log to np.log
Change xr.ufuncs.log to np.log, edits to use of log
Mar 17, 2023
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
I tried using the log target variable option and found that
xr.ufuncs.log
is now outdated and we should switch tonp.log
, as documented here:Line to change.
Also, I think the val and tst data need to be logged here. Why are only the training data logged? The mean and standard deviation are computed on the logged data and then applied to the val and tst data, which are not logged. Some comments in the
preproc_utils
need to be updated to describe whatlog_y_vars
andlog_vars
should be.And we should subtract off the small constant after transforming from log to real space here.
I can address these edits in a PR
The text was updated successfully, but these errors were encountered: