search

ikhlestov / vision_networks

Repo about neural networks for images handling
MIT License
264 stars 122 forks source link

The order of image normalization and augmentation seems to be the same as the original implementation? #14

Closed zhaohuihai closed 6 years ago

zhaohuihai commented 7 years ago

You mentioned that "For Cifar+ datasets image normalization was performed before augmentation. This may cause a little bit lower results than reported in paper." However, I check the image preprocessing parts of these two implementations, and I find that both of them apply normalization before augmentation, so there should be other reason for the difference in performance.

curiositywan commented 6 years ago

I have the same question. I think the question should be whether the order of image normalization and augmentation matter since I notice some other implementations apply normalization after data augmentation during training. If yes, so which order should be better?

ikhlestov commented 6 years ago

Hi all! Here I try to implement code the same way as it was in original repo, but there still can be some difference. Really I don't know exactly what order of augmentation should be correct and better. It should be tested. Unfortunately, I have very tight schedule now. So, if somebody wants to make such experiment and send a result/updated code with PR - it will be really great!

curiositywan commented 6 years ago

I've checked the original repo. There is only one minor difference between yours and the authors'. For the augmentation part, the authors do flipping before translating, as in yours, the order is reverse. That's the only difference regarding data processing.

illarion-rl commented 6 years ago

Ok, I will change the order and re-run the experiments, when I'm free.

ikhlestov commented 6 years ago

Well, for CIFAR100+ dataset I've get an error 25.62 when previous was 25.87. I don't know did this happen because of another augmentation order or because of another initialization. But at least it didn't become worse. So I'm going to change the order. Thank you for reporting the issue!