-
Type:
Task
-
Status: Open (View Workflow)
-
Priority:
Normal
-
Resolution: Unresolved
-
Labels:None
Created on 2025-03-05 16:53:10 by Didier Vibert. % Done: 0
This parameter was needed and used in @DeepLearningSolve@ for being able to apply to some amazed>0.38 outputs a Deep Learning network learned on outputs produced from an amazed version <= 0.38 ie before the implementation of the "compressed" pdf, when amazed switched from the systematic interpolation of the pdf on a fine grid to no interpolation with the PDF stored only on the coarse grid + 2nd pass windows (see #7137).
The difference comes from the fact that the last sample of the zgrid after #7137 is the end of the coarse grid whereas it was the end of the fine grid before (see #7530-9).
TODO:
- check we don't use anymore a network learned from amazed <=0.38 outputs
- remove the @zgrid_max@ param in @PdfHandler.convertToRegular@
- remove @CLogZPdfResult::extrapolate_on_right_border@