search

junjun-jiang / BaMBNet

The official repository for BaMBNet
22 stars 2 forks source link

The input of model in train.py and test.py #6

Closed FlyFish-space closed 2 years ago

FlyFish-space commented 2 years ago
  1. In your code train.py, the model input is (l_img, r_img, b_img). Is b_img COC map of train dataset? But b_img is not obtained in your blur_train.py. So do I need to change your code blur_test.py to obtain b_img?
  2. In your code test.py, the model input is (l_img, r_img, r_img).If there’s been a mistake？Why do you concatnate two r_img?
junjun-jiang commented 2 years ago

During the testing phase, we do not need the COC images any more. The input of evaluation model is just used to reuse code.

FlyFish-space commented 2 years ago

Is b_img in the training code COC Map of training data?

junjun-jiang commented 2 years ago

Yeah. We used an annealing strategy during training. Therefore, the b_img is used for initial several epochs.