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
Currently I am trying to use this framework for image recognition, but after a while I get the exeption "Error during claculation". The sgd trainer increased the gradients until the gradients of one layer reached neagtive infintity as far as I saw. Projekt_NN_Conv_Net.zip
The text was updated successfully, but these errors were encountered:
I think that you are not giving one-hot encoded labels as Y while training.
i.e. LabelVolume should have a shape of [1, 1, NumberOfCategories, BatchSize] like it is done in this Classify2DExample
Currently I am trying to use this framework for image recognition, but after a while I get the exeption "Error during claculation". The sgd trainer increased the gradients until the gradients of one layer reached neagtive infintity as far as I saw.
Projekt_NN_Conv_Net.zip
The text was updated successfully, but these errors were encountered: