-
Notifications
You must be signed in to change notification settings - Fork 32
Crashes when testing #5
Copy link
Copy link
Open
Description
When I did testing on VOC2012 using the pretrained resnet model, I encountered the following error:
seg_accuracy_layer.cpp:87] Unexpected label 220. num: 0. row: 109. col: 291
I tried to modify test.prototxt to ignore label 220, however I again encountered
seg_accuracy_layer.cpp:87] Unexpected label 38. num: 0. row: 109. col: 291
Seems I couldn't just keep adding all problematic labels. A Google search led to a suggestion to replace the "SegAccuracy" layer to an "Accuracy" layer. It indeed fixed the errors, however the final reported accuracy is
fc1_accuracy = 0.460454
which is far below the performance on the VOC leaderboard.
What's wrong here? Thanks for your help.
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
No labels