This repository has been archived by the owner on Nov 16, 2023. It is now read-only.
debug test.py patch padding before model scoring - facilitates scoring correctness and higher scoring accuracy #330
Labels
Prior: Medium
Type: Accuracy
related to increasing performance accuracy
Type: Bug
Something isn't working
Milestone
Images are being fed for scoring which don’t have any useful data or have a large amount of padding – not sure if this is useful – why are we scoring images which are entirely padded, or largely padded?
The text was updated successfully, but these errors were encountered: