Closed yangxh11 closed 4 years ago
did you figure out why? i got the same thing here in 2023 :D
bs8: 0.412 bs128: 0.726 bs256: 0.810
it seems memory is all you need to solve cub-200 lol
@yangze0930 ! do you know why this problem arises?
found the problem. I forgot to set net.eval(). Everything works now.
The model is fully trained and saved. When testing, I only change the BATCH_SIZE in config.py and get different test accuracy.